Daily Papers

We present CLIP2Video network to transfer the image-language pre-training model to video-text retrieval in an end-to-end manner. Leading approaches in the domain of video-and-language learning try to distill the spatio-temporal video features and multi-modal interaction between videos and languages from a large-scale video-text dataset. Different from them, we leverage pretrained image-language model, simplify it as a two-stage framework with co-learning of image-text and enhancing temporal relations between video frames and video-text respectively, make it able to train on comparatively small datasets. Specifically, based on the spatial semantics captured by Contrastive Language-Image Pretraining (CLIP) model, our model involves a Temporal Difference Block to capture motions at fine temporal video frames, and a Temporal Alignment Block to re-align the tokens of video clips and phrases and enhance the multi-modal correlation. We conduct thorough ablation studies, and achieve state-of-the-art performance on major text-to-video and video-to-text retrieval benchmarks, including new records of retrieval accuracy on MSR-VTT, MSVD and VATEX.

Tracking points in videos is typically formulated as frame-to-frame correspondence, where each point is matched locally to the next frame. While this works over short horizons, errors accumulate under articulation, occlusion, and viewpoint change, leading to silent semantic drift that existing trackers cannot detect or correct. In this work, we revisit long-horizon tracking from a monitoring perspective and introduce QueST, a monitoring-by-design framework that treats interaction-relevant entities as persistent semantic queries rather than transient point tracks. Instead of local propagation, each query attends globally over spatio-temporal video features at every time-step, providing a stable semantic anchor across time. We further constrain query trajectories with lightweight 3D physical grounding, using geometric plausibility to suppress unbounded drift under occlusion. We evaluate QueST on long-horizon articulated sequences from PartNet-Mobility in SAPIEN and compare against RAFT-3D, CoTracker, and TAP-Net. QueST substantially reduces terminal drift achieving a 67.7% Absolute Point Error (APE) improvement over TAP-Net while better preserving identity over extended horizons. Our results show that embedding semantic monitoring directly into perception enables more reliable long-horizon tracking under distribution shift.

Detection of anomaly events is relevant for public safety and requires a combination of fine-grained motion information and contextual events at variable time-scales. To this end, we propose a Multi-Timescale Feature Learning (MTFL) method to enhance the representation of anomaly features. Short, medium, and long temporal tubelets are employed to extract spatio-temporal video features using a Video Swin Transformer. Experimental results demonstrate that MTFL outperforms state-of-the-art methods on the UCF-Crime dataset, achieving an anomaly detection performance 89.78% AUC. Moreover, it performs complementary to SotA with 95.32% AUC on the ShanghaiTech and 84.57% AP on the XD-Violence dataset. Furthermore, we generate an extended dataset of the UCF-Crime for development and evaluation on a wider range of anomalies, namely Video Anomaly Detection Dataset (VADD), involving 2,591 videos in 18 classes with extensive coverage of realistic anomalies.

We propose Spatio-temporal Crop Aggregation for video representation LEarning (SCALE), a novel method that enjoys high scalability at both training and inference time. Our model builds long-range video features by learning from sets of video clip-level features extracted with a pre-trained backbone. To train the model, we propose a self-supervised objective consisting of masked clip feature prediction. We apply sparsity to both the input, by extracting a random set of video clips, and to the loss function, by only reconstructing the sparse inputs. Moreover, we use dimensionality reduction by working in the latent space of a pre-trained backbone applied to single video clips. These techniques make our method not only extremely efficient to train but also highly effective in transfer learning. We demonstrate that our video representation yields state-of-the-art performance with linear, non-linear, and KNN probing on common action classification and video understanding datasets.

Frame quality deterioration is one of the main challenges in the field of video understanding. To compensate for the information loss caused by deteriorated frames, recent approaches exploit transformer-based integration modules to obtain spatio-temporal information. However, these integration modules are heavy and complex. Furthermore, each integration module is specifically tailored for its target task, making it difficult to generalise to multiple tasks. In this paper, we present a neat and unified framework, called Spatio-Temporal Prompting Network (STPN). It can efficiently extract robust and accurate video features by dynamically adjusting the input features in the backbone network. Specifically, STPN predicts several video prompts containing spatio-temporal information of neighbour frames. Then, these video prompts are prepended to the patch embeddings of the current frame as the updated input for video feature extraction. Moreover, STPN is easy to generalise to various video tasks because it does not contain task-specific modules. Without bells and whistles, STPN achieves state-of-the-art performance on three widely-used datasets for different video understanding tasks, i.e., ImageNetVID for video object detection, YouTubeVIS for video instance segmentation, and GOT-10k for visual object tracking. Code is available at https://github.com/guanxiongsun/vfe.pytorch.

Referring Video Object Segmentation (RVOS) aims to segment target objects in videos based on natural language descriptions. However, fixed keyframe-based approaches that couple a vision language model with a separate propagation module often fail to capture rapidly changing spatiotemporal dynamics and to handle queries requiring multi-step reasoning, leading to sharp performance drops on motion-intensive and reasoning-oriented videos beyond static RVOS benchmarks. To address these limitations, we propose VIRST (Video-Instructed Reasoning Assistant for Spatio-Temporal Segmentation), an end-to-end framework that unifies global video reasoning and pixel-level mask prediction within a single model. VIRST bridges semantic and segmentation representations through the Spatio-Temporal Fusion (STF), which fuses segmentation-aware video features into the vision-language backbone, and employs the Temporal Dynamic Anchor Updater to maintain temporally adjacent anchor frames that provide stable temporal cues under large motion, occlusion, and reappearance. This unified design achieves state-of-the-art results across diverse RVOS benchmarks under realistic and challenging conditions, demonstrating strong generalization to both referring and reasoning oriented settings. The code and checkpoints are available at https://github.com/AIDASLab/VIRST.

One of the main challenges of multimodal learning is the need to combine heterogeneous modalities (e.g., video, audio, text). For example, video and audio are obtained at much higher rates than text and are roughly aligned in time. They are often not synchronized with text, which comes as a global context, e.g., a title, or a description. Furthermore, video and audio inputs are of much larger volumes, and grow as the video length increases, which naturally requires more compute dedicated to these modalities and makes modeling of long-range dependencies harder. We here decouple the multimodal modeling, dividing it into separate, focused autoregressive models, processing the inputs according to the characteristics of the modalities. We propose a multimodal model, called Mirasol3B, consisting of an autoregressive component for the time-synchronized modalities (audio and video), and an autoregressive component for the context modalities which are not necessarily aligned in time but are still sequential. To address the long-sequences of the video-audio inputs, we propose to further partition the video and audio sequences in consecutive snippets and autoregressively process their representations. To that end, we propose a Combiner mechanism, which models the audio-video information jointly within a timeframe. The Combiner learns to extract audio and video features from raw spatio-temporal signals, and then learns to fuse these features producing compact but expressive representations per snippet. Our approach achieves the state-of-the-art on well established multimodal benchmarks, outperforming much larger models. It effectively addresses the high computational demand of media inputs by both learning compact representations, controlling the sequence length of the audio-video feature representations, and modeling their dependencies in time.

We introduce TAPIP3D, a novel approach for long-term 3D point tracking in monocular RGB and RGB-D videos. TAPIP3D represents videos as camera-stabilized spatio-temporal feature clouds, leveraging depth and camera motion information to lift 2D video features into a 3D world space where camera motion is effectively canceled. TAPIP3D iteratively refines multi-frame 3D motion estimates within this stabilized representation, enabling robust tracking over extended periods. To manage the inherent irregularities of 3D point distributions, we propose a Local Pair Attention mechanism. This 3D contextualization strategy effectively exploits spatial relationships in 3D, forming informative feature neighborhoods for precise 3D trajectory estimation. Our 3D-centric approach significantly outperforms existing 3D point tracking methods and even enhances 2D tracking accuracy compared to conventional 2D pixel trackers when accurate depth is available. It supports inference in both camera coordinates (i.e., unstabilized) and world coordinates, and our results demonstrate that compensating for camera motion improves tracking performance. Our approach replaces the conventional 2D square correlation neighborhoods used in prior 2D and 3D trackers, leading to more robust and accurate results across various 3D point tracking benchmarks. Project Page: https://tapip3d.github.io

This paper strives to solve complex video question answering (VideoQA) which features long video containing multiple objects and events at different time. To tackle the challenge, we highlight the importance of identifying question-critical temporal moments and spatial objects from the vast amount of video content. Towards this, we propose a Spatio-Temporal Rationalization (STR), a differentiable selection module that adaptively collects question-critical moments and objects using cross-modal interaction. The discovered video moments and objects are then served as grounded rationales to support answer reasoning. Based on STR, we further propose TranSTR, a Transformer-style neural network architecture that takes STR as the core and additionally underscores a novel answer interaction mechanism to coordinate STR for answer decoding. Experiments on four datasets show that TranSTR achieves new state-of-the-art (SoTA). Especially, on NExT-QA and Causal-VidQA which feature complex VideoQA, it significantly surpasses the previous SoTA by 5.8\% and 6.8\%, respectively. We then conduct extensive studies to verify the importance of STR as well as the proposed answer interaction mechanism. With the success of TranSTR and our comprehensive analysis, we hope this work can spark more future efforts in complex VideoQA. Code will be released at https://github.com/yl3800/TranSTR.

We address the problem of video representation learning without human-annotated labels. While previous efforts address the problem by designing novel self-supervised tasks using video data, the learned features are merely on a frame-by-frame basis, which are not applicable to many video analytic tasks where spatio-temporal features are prevailing. In this paper we propose a novel self-supervised approach to learn spatio-temporal features for video representation. Inspired by the success of two-stream approaches in video classification, we propose to learn visual features by regressing both motion and appearance statistics along spatial and temporal dimensions, given only the input video data. Specifically, we extract statistical concepts (fast-motion region and the corresponding dominant direction, spatio-temporal color diversity, dominant color, etc.) from simple patterns in both spatial and temporal domains. Unlike prior puzzles that are even hard for humans to solve, the proposed approach is consistent with human inherent visual habits and therefore easy to answer. We conduct extensive experiments with C3D to validate the effectiveness of our proposed approach. The experiments show that our approach can significantly improve the performance of C3D when applied to video classification tasks. Code is available at https://github.com/laura-wang/video_repres_mas.

The spread of the Coronavirus disease-2019 epidemic has caused many courses and exams to be conducted online. The cheating behavior detection model in examination invigilation systems plays a pivotal role in guaranteeing the equality of long-distance examinations. However, cheating behavior is rare, and most researchers do not comprehensively take into account features such as head posture, gaze angle, body posture, and background information in the task of cheating behavior detection. In this paper, we develop and present CHEESE, a CHEating detection framework via multiplE inStancE learning. The framework consists of a label generator that implements weak supervision and a feature encoder to learn discriminative features. In addition, the framework combines body posture and background features extracted by 3D convolution with eye gaze, head posture and facial features captured by OpenFace 2.0. These features are fed into the spatio-temporal graph module by stitching to analyze the spatio-temporal changes in video clips to detect the cheating behaviors. Our experiments on three datasets, UCF-Crime, ShanghaiTech and Online Exam Proctoring (OEP), prove the effectiveness of our method as compared to the state-of-the-art approaches, and obtain the frame-level AUC score of 87.58% on the OEP dataset.

This paper introduces ViNet-S, a 36MB model based on the ViNet architecture with a U-Net design, featuring a lightweight decoder that significantly reduces model size and parameters without compromising performance. Additionally, ViNet-A (148MB) incorporates spatio-temporal action localization (STAL) features, differing from traditional video saliency models that use action classification backbones. Our studies show that an ensemble of ViNet-S and ViNet-A, by averaging predicted saliency maps, achieves state-of-the-art performance on three visual-only and six audio-visual saliency datasets, outperforming transformer-based models in both parameter efficiency and real-time performance, with ViNet-S reaching over 1000fps.

Current weakly supervised video anomaly detection (WSVAD) task aims to achieve frame-level anomalous event detection with only coarse video-level annotations available. Existing works typically involve extracting global features from full-resolution video frames and training frame-level classifiers to detect anomalies in the temporal dimension. However, most anomalous events tend to occur in localized spatial regions rather than the entire video frames, which implies existing frame-level feature based works may be misled by the dominant background information and lack the interpretation of the detected anomalies. To address this dilemma, this paper introduces a novel method called STPrompt that learns spatio-temporal prompt embeddings for weakly supervised video anomaly detection and localization (WSVADL) based on pre-trained vision-language models (VLMs). Our proposed method employs a two-stream network structure, with one stream focusing on the temporal dimension and the other primarily on the spatial dimension. By leveraging the learned knowledge from pre-trained VLMs and incorporating natural motion priors from raw videos, our model learns prompt embeddings that are aligned with spatio-temporal regions of videos (e.g., patches of individual frames) for identify specific local regions of anomalies, enabling accurate video anomaly detection while mitigating the influence of background information. Without relying on detailed spatio-temporal annotations or auxiliary object detection/tracking, our method achieves state-of-the-art performance on three public benchmarks for the WSVADL task.

Video Anomaly Detection (VAD) presents a significant challenge in computer vision, particularly due to the unpredictable and infrequent nature of anomalous events, coupled with the diverse and dynamic environments in which they occur. Human-centric VAD, a specialized area within this domain, faces additional complexities, including variations in human behavior, potential biases in data, and substantial privacy concerns related to human subjects. These issues complicate the development of models that are both robust and generalizable. To address these challenges, recent advancements have focused on pose-based VAD, which leverages human pose as a high-level feature to mitigate privacy concerns, reduce appearance biases, and minimize background interference. In this paper, we introduce SPARTA, a novel transformer-based architecture designed specifically for human-centric pose-based VAD. SPARTA introduces an innovative Spatio-Temporal Pose and Relative Pose (ST-PRP) tokenization method that produces an enriched representation of human motion over time. This approach ensures that the transformer's attention mechanism captures both spatial and temporal patterns simultaneously, rather than focusing on only one aspect. The addition of the relative pose further emphasizes subtle deviations from normal human movements. The architecture's core, a novel Unified Encoder Twin Decoders (UETD) transformer, significantly improves the detection of anomalous behaviors in video data. Extensive evaluations across multiple benchmark datasets demonstrate that SPARTA consistently outperforms existing methods, establishing a new state-of-the-art in pose-based VAD.

Neural Video Compression has emerged in recent years, with condition-based frameworks outperforming traditional codecs. However, most existing methods rely solely on the previous frame's features to predict temporal context, leading to two critical issues. First, the short reference window misses long-term dependencies and fine texture details. Second, propagating only feature-level information accumulates errors over frames, causing prediction inaccuracies and loss of subtle textures. To address these, we propose the Long-term Spatio-Temporal Enhanced Context (L-STEC) method. We first extend the reference chain with LSTM to capture long-term dependencies. We then incorporate warped spatial context from the pixel domain, fusing spatio-temporal information through a multi-receptive field network to better preserve reference details. Experimental results show that L-STEC significantly improves compression by enriching contextual information, achieving 37.01% bitrate savings in PSNR and 31.65% in MS-SSIM compared to DCVC-TCM, outperforming both VTM-17.0 and DCVC-FM and establishing new state-of-the-art performance.

Text-to-image (T2I) diffusion models achieve state-of-the-art results in image synthesis and editing. However, leveraging such pretrained models for video editing is considered a major challenge. Many existing works attempt to enforce temporal consistency in the edited video through explicit correspondence mechanisms, either in pixel space or between deep features. These methods, however, struggle with strong nonrigid motion. In this paper, we introduce a fundamentally different approach, which is based on the observation that spatiotemporal slices of natural videos exhibit similar characteristics to natural images. Thus, the same T2I diffusion model that is normally used only as a prior on video frames, can also serve as a strong prior for enhancing temporal consistency by applying it on spatiotemporal slices. Based on this observation, we present Slicedit, a method for text-based video editing that utilizes a pretrained T2I diffusion model to process both spatial and spatiotemporal slices. Our method generates videos that retain the structure and motion of the original video while adhering to the target text. Through extensive experiments, we demonstrate Slicedit's ability to edit a wide range of real-world videos, confirming its clear advantages compared to existing competing methods. Webpage: https://matankleiner.github.io/slicedit/

The application of video captioning models aims at translating the content of videos by using accurate natural language. Due to the complex nature inbetween object interaction in the video, the comprehensive understanding of spatio-temporal relations of objects remains a challenging task. Existing methods often fail in generating sufficient feature representations of video content. In this paper, we propose a video captioning model based on dual graphs and gated fusion: we adapt two types of graphs to generate feature representations of video content and utilize gated fusion to further understand these different levels of information. Using a dual-graphs model to generate appearance features and motion features respectively can utilize the content correlation in frames to generate various features from multiple perspectives. Among them, dual-graphs reasoning can enhance the content correlation in frame sequences to generate advanced semantic features; The gated fusion, on the other hand, aggregates the information in multiple feature representations for comprehensive video content understanding. The experiments conducted on worldly used datasets MSVD and MSR-VTT demonstrate state-of-the-art performance of our proposed approach.

Video Captioning and Summarization have become very popular in the recent years due to advancements in Sequence Modelling, with the resurgence of Long-Short Term Memory networks (LSTMs) and introduction of Gated Recurrent Units (GRUs). Existing architectures extract spatio-temporal features using CNNs and utilize either GRUs or LSTMs to model dependencies with soft attention layers. These attention layers do help in attending to the most prominent features and improve upon the recurrent units, however, these models suffer from the inherent drawbacks of the recurrent units themselves. The introduction of the Transformer model has driven the Sequence Modelling field into a new direction. In this project, we implement a Transformer-based model for Video captioning, utilizing 3D CNN architectures like C3D and Two-stream I3D for video extraction. We also apply certain dimensionality reduction techniques so as to keep the overall size of the model within limits. We finally present our results on the MSVD and ActivityNet datasets for Single and Dense video captioning tasks respectively.

Research on video frame interpolation has made significant progress in recent years. However, existing methods mostly use off-the-shelf metrics to measure the quality of interpolation results with the exception of a few methods that employ user studies, which is time-consuming. As video frame interpolation results often exhibit unique artifacts, existing quality metrics sometimes are not consistent with human perception when measuring the interpolation results. Some recent deep learning-based perceptual quality metrics are shown more consistent with human judgments, but their performance on videos is compromised since they do not consider temporal information. In this paper, we present a dedicated perceptual quality metric for measuring video frame interpolation results. Our method learns perceptual features directly from videos instead of individual frames. It compares pyramid features extracted from video frames and employs Swin Transformer blocks-based spatio-temporal modules to extract spatio-temporal information. To train our metric, we collected a new video frame interpolation quality assessment dataset. Our experiments show that our dedicated quality metric outperforms state-of-the-art methods when measuring video frame interpolation results. Our code and model are made publicly available at https://github.com/hqqxyy/VFIPS.

Pure vision transformer architectures are highly effective for short video classification and action recognition tasks. However, due to the quadratic complexity of self attention and lack of inductive bias, transformers are resource intensive and suffer from data inefficiencies. Long form video understanding tasks amplify data and memory efficiency problems in transformers making current approaches unfeasible to implement on data or memory restricted domains. This paper introduces an efficient Spatio-Temporal Attention Network (STAN) which uses a two-stream transformer architecture to model dependencies between static image features and temporal contextual features. Our proposed approach can classify videos up to two minutes in length on a single GPU, is data efficient, and achieves SOTA performance on several long video understanding tasks.

The rapid growth of user-generated (video) content (UGC) has driven increased demand for research on no-reference (NR) perceptual video quality assessment (VQA). NR-VQA is a key component for large-scale video quality monitoring in social media and streaming applications where a pristine reference is not available. This paper proposes a novel NR-VQA model based on spatio-temporal fragmentation driven by inter-frame variations. By leveraging these inter-frame differences, the model progressively analyses quality-sensitive regions at multiple levels: frames, patches, and fragmented frames. It integrates frames, fragmented residuals, and fragmented frames aligned with residuals to effectively capture global and local information. The model extracts both 2D and 3D features in order to characterize these spatio-temporal variations. Experiments conducted on five UGC datasets and against state-of-the-art models ranked our proposed method among the top 2 in terms of average rank correlation (DIVA-VQA-L: 0.898 and DIVA-VQA-B: 0.886). The improved performance is offered at a low runtime complexity, with DIVA-VQA-B ranked top and DIVA-VQA-L third on average compared to the fastest existing NR-VQA method. Code and models are publicly available at: https://github.com/xinyiW915/DIVA-VQA.

Video object removal and inpainting are critical tasks in the fields of computer vision and multimedia processing, aimed at restoring missing or corrupted regions in video sequences. Traditional methods predominantly rely on flow-based propagation and spatio-temporal Transformers, but these approaches face limitations in effectively leveraging long-term temporal features and ensuring temporal consistency in the completion results, particularly when dealing with large masks. Consequently, performance on extensive masked areas remains suboptimal. To address these challenges, this paper introduces a novel video inpainting approach leveraging the Diffusion Transformer (DiT). DiT synergistically combines the advantages of diffusion models and transformer architectures to maintain long-term temporal consistency while ensuring high-quality inpainting results. We propose a Circular Position-Shift strategy to further enhance long-term temporal consistency during the inference stage. Additionally, the proposed method interactively removes specified objects, and generates corresponding prompts. In terms of processing speed, it takes only 65 seconds (testing on one NVIDIA H800 GPU) to complete a video with a resolution of 2160 times 2100 with 97 frames without any acceleration method. Experimental results indicate that the proposed method demonstrates superior performance in content fidelity, texture restoration, and temporal consistency. Project page:https://jieliu95.github.io/EraserDiT_demo/

With the rapid growth of User-Generated Content (UGC) exchanged between users and sharing platforms, the need for video quality assessment in the wild is increasingly evident. UGC is typically acquired using consumer devices and undergoes multiple rounds of compression (transcoding) before reaching the end user. Therefore, traditional quality metrics that employ the original content as a reference are not suitable. In this paper, we propose ReLaX-VQA, a novel No-Reference Video Quality Assessment (NR-VQA) model that aims to address the challenges of evaluating the quality of diverse video content without reference to the original uncompressed videos. ReLaX-VQA uses frame differences to select spatio-temporal fragments intelligently together with different expressions of spatial features associated with the sampled frames. These are then used to better capture spatial and temporal variabilities in the quality of neighbouring frames. Furthermore, the model enhances abstraction by employing layer-stacking techniques in deep neural network features from Residual Networks and Vision Transformers. Extensive testing across four UGC datasets demonstrates that ReLaX-VQA consistently outperforms existing NR-VQA methods, achieving an average SRCC of 0.8658 and PLCC of 0.8873. Open-source code and trained models that will facilitate further research and applications of NR-VQA can be found at https://github.com/xinyiW915/ReLaX-VQA.

We propose MAMo, a novel memory and attention frame-work for monocular video depth estimation. MAMo can augment and improve any single-image depth estimation networks into video depth estimation models, enabling them to take advantage of the temporal information to predict more accurate depth. In MAMo, we augment model with memory which aids the depth prediction as the model streams through the video. Specifically, the memory stores learned visual and displacement tokens of the previous time instances. This allows the depth network to cross-reference relevant features from the past when predicting depth on the current frame. We introduce a novel scheme to continuously update the memory, optimizing it to keep tokens that correspond with both the past and the present visual information. We adopt attention-based approach to process memory features where we first learn the spatio-temporal relation among the resultant visual and displacement memory tokens using self-attention module. Further, the output features of self-attention are aggregated with the current visual features through cross-attention. The cross-attended features are finally given to a decoder to predict depth on the current frame. Through extensive experiments on several benchmarks, including KITTI, NYU-Depth V2, and DDAD, we show that MAMo consistently improves monocular depth estimation networks and sets new state-of-the-art (SOTA) accuracy. Notably, our MAMo video depth estimation provides higher accuracy with lower latency, when omparing to SOTA cost-volume-based video depth models.

We present the Decoupled VIdeo Segmentation (DVIS) framework, a novel approach for the challenging task of universal video segmentation, including video instance segmentation (VIS), video semantic segmentation (VSS), and video panoptic segmentation (VPS). Unlike previous methods that model video segmentation in an end-to-end manner, our approach decouples video segmentation into three cascaded sub-tasks: segmentation, tracking, and refinement. This decoupling design allows for simpler and more effective modeling of the spatio-temporal representations of objects, especially in complex scenes and long videos. Accordingly, we introduce two novel components: the referring tracker and the temporal refiner. These components track objects frame by frame and model spatio-temporal representations based on pre-aligned features. To improve the tracking capability of DVIS, we propose a denoising training strategy and introduce contrastive learning, resulting in a more robust framework named DVIS++. Furthermore, we evaluate DVIS++ in various settings, including open vocabulary and using a frozen pre-trained backbone. By integrating CLIP with DVIS++, we present OV-DVIS++, the first open-vocabulary universal video segmentation framework. We conduct extensive experiments on six mainstream benchmarks, including the VIS, VSS, and VPS datasets. Using a unified architecture, DVIS++ significantly outperforms state-of-the-art specialized methods on these benchmarks in both close- and open-vocabulary settings. Code:~https://github.com/zhang-tao-whu/DVIS_Plus.

Existing multimodal-based human action recognition approaches are computationally intensive, limiting their deployment in real-time applications. In this work, we present a novel and efficient pose-driven attention-guided multimodal network (EPAM-Net) for action recognition in videos. Specifically, we propose eXpand temporal Shift (X-ShiftNet) convolutional architectures for RGB and pose streams to capture spatio-temporal features from RGB videos and their skeleton sequences. The X-ShiftNet tackles the high computational cost of the 3D CNNs by integrating the Temporal Shift Module (TSM) into an efficient 2D CNN, enabling efficient spatiotemporal learning. Then skeleton features are utilized to guide the visual network stream, focusing on keyframes and their salient spatial regions using the proposed spatial-temporal attention block. Finally, the predictions of the two streams are fused for final classification. The experimental results show that our method, with a significant reduction in floating-point operations (FLOPs), outperforms and competes with the state-of-the-art methods on NTU RGB-D 60, NTU RGB-D 120, PKU-MMD, and Toyota SmartHome datasets. The proposed EPAM-Net provides up to a 72.8x reduction in FLOPs and up to a 48.6x reduction in the number of network parameters. The code will be available at https://github.com/ahmed-nady/Multimodal-Action-Recognition.

Weakly-supervised methods for video anomaly detection (VAD) are conventionally based merely on RGB spatio-temporal features, which continues to limit their reliability in real-world scenarios. This is due to the fact that RGB-features are not sufficiently distinctive in setting apart categories such as shoplifting from visually similar events. Therefore, towards robust complex real-world VAD, it is essential to augment RGB spatio-temporal features by additional modalities. Motivated by this, we introduce the Poly-modal Induced framework for VAD: "PI-VAD", a novel approach that augments RGB representations by five additional modalities. Specifically, the modalities include sensitivity to fine-grained motion (Pose), three dimensional scene and entity representation (Depth), surrounding objects (Panoptic masks), global motion (optical flow), as well as language cues (VLM). Each modality represents an axis of a polygon, streamlined to add salient cues to RGB. PI-VAD includes two plug-in modules, namely Pseudo-modality Generation module and Cross Modal Induction module, which generate modality-specific prototypical representation and, thereby, induce multi-modal information into RGB cues. These modules operate by performing anomaly-aware auxiliary tasks and necessitate five modality backbones -- only during training. Notably, PI-VAD achieves state-of-the-art accuracy on three prominent VAD datasets encompassing real-world scenarios, without requiring the computational overhead of five modality backbones at inference.

This paper proposes a self-supervised approach to learn universal facial representations from videos, that can transfer across a variety of facial analysis tasks such as Facial Attribute Recognition (FAR), Facial Expression Recognition (FER), DeepFake Detection (DFD), and Lip Synchronization (LS). Our proposed framework, named MARLIN, is a facial video masked autoencoder, that learns highly robust and generic facial embeddings from abundantly available non-annotated web crawled facial videos. As a challenging auxiliary task, MARLIN reconstructs the spatio-temporal details of the face from the densely masked facial regions which mainly include eyes, nose, mouth, lips, and skin to capture local and global aspects that in turn help in encoding generic and transferable features. Through a variety of experiments on diverse downstream tasks, we demonstrate MARLIN to be an excellent facial video encoder as well as feature extractor, that performs consistently well across a variety of downstream tasks including FAR (1.13% gain over supervised benchmark), FER (2.64% gain over unsupervised benchmark), DFD (1.86% gain over unsupervised benchmark), LS (29.36% gain for Frechet Inception Distance), and even in low data regime. Our code and models are available at https://github.com/ControlNet/MARLIN .

In recent years, significant progress has been made in video instance segmentation (VIS), with many offline and online methods achieving state-of-the-art performance. While offline methods have the advantage of producing temporally consistent predictions, they are not suitable for real-time scenarios. Conversely, online methods are more practical, but maintaining temporal consistency remains a challenging task. In this paper, we propose a novel online method for video instance segmentation, called TCOVIS, which fully exploits the temporal information in a video clip. The core of our method consists of a global instance assignment strategy and a spatio-temporal enhancement module, which improve the temporal consistency of the features from two aspects. Specifically, we perform global optimal matching between the predictions and ground truth across the whole video clip, and supervise the model with the global optimal objective. We also capture the spatial feature and aggregate it with the semantic feature between frames, thus realizing the spatio-temporal enhancement. We evaluate our method on four widely adopted VIS benchmarks, namely YouTube-VIS 2019/2021/2022 and OVIS, and achieve state-of-the-art performance on all benchmarks without bells-and-whistles. For instance, on YouTube-VIS 2021, TCOVIS achieves 49.5 AP and 61.3 AP with ResNet-50 and Swin-L backbones, respectively. Code is available at https://github.com/jun-long-li/TCOVIS.

Text-to-video (T2V) generative models have rapidly advanced and found widespread applications across fields like entertainment, education, and marketing. However, the adversarial vulnerabilities of these models remain rarely explored. We observe that in T2V generation tasks, the generated videos often contain substantial redundant information not explicitly specified in the text prompts, such as environmental elements, secondary objects, and additional details, providing opportunities for malicious attackers to embed hidden harmful content. Exploiting this inherent redundancy, we introduce BadVideo, the first backdoor attack framework tailored for T2V generation. Our attack focuses on designing target adversarial outputs through two key strategies: (1) Spatio-Temporal Composition, which combines different spatiotemporal features to encode malicious information; (2) Dynamic Element Transformation, which introduces transformations in redundant elements over time to convey malicious information. Based on these strategies, the attacker's malicious target seamlessly integrates with the user's textual instructions, providing high stealthiness. Moreover, by exploiting the temporal dimension of videos, our attack successfully evades traditional content moderation systems that primarily analyze spatial information within individual frames. Extensive experiments demonstrate that BadVideo achieves high attack success rates while preserving original semantics and maintaining excellent performance on clean inputs. Overall, our work reveals the adversarial vulnerability of T2V models, calling attention to potential risks and misuse. Our project page is at https://wrt2000.github.io/BadVideo2025/.

In this paper, we propose a simple yet effective approach for self-supervised video object segmentation (VOS). Our key insight is that the inherent structural dependencies present in DINO-pretrained Transformers can be leveraged to establish robust spatio-temporal correspondences in videos. Furthermore, simple clustering on this correspondence cue is sufficient to yield competitive segmentation results. Previous self-supervised VOS techniques majorly resort to auxiliary modalities or utilize iterative slot attention to assist in object discovery, which restricts their general applicability and imposes higher computational requirements. To deal with these challenges, we develop a simplified architecture that capitalizes on the emerging objectness from DINO-pretrained Transformers, bypassing the need for additional modalities or slot attention. Specifically, we first introduce a single spatio-temporal Transformer block to process the frame-wise DINO features and establish spatio-temporal dependencies in the form of self-attention. Subsequently, utilizing these attention maps, we implement hierarchical clustering to generate object segmentation masks. To train the spatio-temporal block in a fully self-supervised manner, we employ semantic and dynamic motion consistency coupled with entropy normalization. Our method demonstrates state-of-the-art performance across multiple unsupervised VOS benchmarks and particularly excels in complex real-world multi-object video segmentation tasks such as DAVIS-17-Unsupervised and YouTube-VIS-19. The code and model checkpoints will be released at https://github.com/shvdiwnkozbw/SSL-UVOS.

Text-to-video (T2V) synthesis has gained increasing attention in the community, in which the recently emerged diffusion models (DMs) have promisingly shown stronger performance than the past approaches. While existing state-of-the-art DMs are competent to achieve high-resolution video generation, they may largely suffer from key limitations (e.g., action occurrence disorders, crude video motions) with respect to the intricate temporal dynamics modeling, one of the crux of video synthesis. In this work, we investigate strengthening the awareness of video dynamics for DMs, for high-quality T2V generation. Inspired by human intuition, we design an innovative dynamic scene manager (dubbed as Dysen) module, which includes (step-1) extracting from input text the key actions with proper time-order arrangement, (step-2) transforming the action schedules into the dynamic scene graph (DSG) representations, and (step-3) enriching the scenes in the DSG with sufficient and reasonable details. Taking advantage of the existing powerful LLMs (e.g., ChatGPT) via in-context learning, Dysen realizes (nearly) human-level temporal dynamics understanding. Finally, the resulting video DSG with rich action scene details is encoded as fine-grained spatio-temporal features, integrated into the backbone T2V DM for video generating. Experiments on popular T2V datasets suggest that our framework consistently outperforms prior arts with significant margins, especially in the scenario with complex actions. Project page at https://haofei.vip/Dysen-VDM

Text-to-video (T2V) diffusion models have shown promising capabilities in synthesizing realistic videos from input text prompts. However, the input text description alone provides limited control over the precise objects movements and camera framing. In this work, we tackle the motion customization problem, where a reference video is provided as motion guidance. While most existing methods choose to fine-tune pre-trained diffusion models to reconstruct the frame differences of the reference video, we observe that such strategy suffer from content leakage from the reference video, and they cannot capture complex motion accurately. To address this issue, we propose MotionMatcher, a motion customization framework that fine-tunes the pre-trained T2V diffusion model at the feature level. Instead of using pixel-level objectives, MotionMatcher compares high-level, spatio-temporal motion features to fine-tune diffusion models, ensuring precise motion learning. For the sake of memory efficiency and accessibility, we utilize a pre-trained T2V diffusion model, which contains considerable prior knowledge about video motion, to compute these motion features. In our experiments, we demonstrate state-of-the-art motion customization performances, validating the design of our framework.

Online video segmentation methods excel at handling long sequences and capturing gradual changes, making them ideal for real-world applications. However, achieving temporally consistent predictions remains a challenge, especially with gradual accumulation of noise or drift in on-line propagation, abrupt occlusions and scene transitions. This paper introduces Local2Global, an online framework, for video instance segmentation, exhibiting state-of-the-art performance with simple baseline and training purely in online fashion. Leveraging the DETR-based query propagation framework, we introduce two novel sets of queries:(1) local queries that capture initial object-specific spatial features from each frame and (2) global queries containing past spatio-temporal representations. We propose the L2G-aligner, a novel lightweight transformer decoder, to facilitate an early alignment between local and global queries. This alignment allows our model to effectively utilize current frame information while maintaining temporal consistency, producing a smooth transition between frames. Furthermore, L2G-aligner is integrated within the segmentation model, without relying on additional complex heuristics, or memory mechanisms. Extensive experiments across various challenging VIS and VPS datasets showcase the superiority of our method with simple online training, surpassing current benchmarks without bells and rings. For instance, we achieve 54.3 and 49.4 AP on Youtube-VIS-19/-21 datasets and 37.0 AP on OVIS dataset respectively withthe ResNet-50 backbone.

Detecting abnormal activities in real-world surveillance videos is an important yet challenging task as the prior knowledge about video anomalies is usually limited or unavailable. Despite that many approaches have been developed to resolve this problem, few of them can capture the normal spatio-temporal patterns effectively and efficiently. Moreover, existing works seldom explicitly consider the local consistency at frame level and global coherence of temporal dynamics in video sequences. To this end, we propose Convolutional Transformer based Dual Discriminator Generative Adversarial Networks (CT-D2GAN) to perform unsupervised video anomaly detection. Specifically, we first present a convolutional transformer to perform future frame prediction. It contains three key components, i.e., a convolutional encoder to capture the spatial information of the input video clips, a temporal self-attention module to encode the temporal dynamics, and a convolutional decoder to integrate spatio-temporal features and predict the future frame. Next, a dual discriminator based adversarial training procedure, which jointly considers an image discriminator that can maintain the local consistency at frame-level and a video discriminator that can enforce the global coherence of temporal dynamics, is employed to enhance the future frame prediction. Finally, the prediction error is used to identify abnormal video frames. Thoroughly empirical studies on three public video anomaly detection datasets, i.e., UCSD Ped2, CUHK Avenue, and Shanghai Tech Campus, demonstrate the effectiveness of the proposed adversarial spatio-temporal modeling framework.

Shape assembly is a ubiquitous task in daily life, integral for constructing complex 3D structures like IKEA furniture. While significant progress has been made in developing autonomous agents for shape assembly, existing datasets have not yet tackled the 4D grounding of assembly instructions in videos, essential for a holistic understanding of assembly in 3D space over time. We introduce IKEA Video Manuals, a dataset that features 3D models of furniture parts, instructional manuals, assembly videos from the Internet, and most importantly, annotations of dense spatio-temporal alignments between these data modalities. To demonstrate the utility of IKEA Video Manuals, we present five applications essential for shape assembly: assembly plan generation, part-conditioned segmentation, part-conditioned pose estimation, video object segmentation, and furniture assembly based on instructional video manuals. For each application, we provide evaluation metrics and baseline methods. Through experiments on our annotated data, we highlight many challenges in grounding assembly instructions in videos to improve shape assembly, including handling occlusions, varying viewpoints, and extended assembly sequences.

Despite rapid advances in 3D content generation, quality assessment for the generated 3D assets remains challenging. Existing methods mainly rely on image-based metrics and operate solely at the object level, limiting their ability to capture spatial coherence, material authenticity, and high-fidelity local details. 1) To address these challenges, we introduce Hi3DEval, a hierarchical evaluation framework tailored for 3D generative content. It combines both object-level and part-level evaluation, enabling holistic assessments across multiple dimensions as well as fine-grained quality analysis. Additionally, we extend texture evaluation beyond aesthetic appearance by explicitly assessing material realism, focusing on attributes such as albedo, saturation, and metallicness. 2) To support this framework, we construct Hi3DBench, a large-scale dataset comprising diverse 3D assets and high-quality annotations, accompanied by a reliable multi-agent annotation pipeline. We further propose a 3D-aware automated scoring system based on hybrid 3D representations. Specifically, we leverage video-based representations for object-level and material-subject evaluations to enhance modeling of spatio-temporal consistency and employ pretrained 3D features for part-level perception. Extensive experiments demonstrate that our approach outperforms existing image-based metrics in modeling 3D characteristics and achieves superior alignment with human preference, providing a scalable alternative to manual evaluations. The project page is available at https://zyh482.github.io/Hi3DEval/.

We describe a novel cross-modal embedding space for actions, named Action2Vec, which combines linguistic cues from class labels with spatio-temporal features derived from video clips. Our approach uses a hierarchical recurrent network to capture the temporal structure of video features. We train our embedding using a joint loss that combines classification accuracy with similarity to Word2Vec semantics. We evaluate Action2Vec by performing zero shot action recognition and obtain state of the art results on three standard datasets. In addition, we present two novel analogy tests which quantify the extent to which our joint embedding captures distributional semantics. This is the first joint embedding space to combine verbs and action videos, and the first to be thoroughly evaluated with respect to its distributional semantics.

Environment understanding in egocentric videos is an important step for applications like robotics, augmented reality and assistive technologies. These videos are characterized by dynamic interactions and a strong dependence on the wearer engagement with the environment. Traditional approaches often focus on isolated clips or fail to integrate rich semantic and geometric information, limiting scene comprehension. We introduce Dynamic Image-Video Feature Fields (DIV FF), a framework that decomposes the egocentric scene into persistent, dynamic, and actor based components while integrating both image and video language features. Our model enables detailed segmentation, captures affordances, understands the surroundings and maintains consistent understanding over time. DIV-FF outperforms state-of-the-art methods, particularly in dynamically evolving scenarios, demonstrating its potential to advance long term, spatio temporal scene understanding.

Human action recognition in long-term videos, characterized by complex backgrounds and subtle action differences, poses significant challenges for traditional deep learning models due to computational overhead, difficulty in capturing long-range temporal dependencies, and limited semantic understanding. While Large Language Models (LLMs) and Large Vision-Language Models (LVLMs) have shown remarkable capabilities in multi-modal understanding and reasoning, their direct application to continuous video streams for fine-grained action recognition remains an open problem. This paper introduces VT-LVLM-AR (Video-Temporal Large Vision-Language Model Adapter for Action Recognition), a novel framework designed to bridge this gap. VT-LVLM-AR comprises a Video-to-Event Mapper (VTEM) that efficiently transforms raw video into compact, semantically rich, and temporally coherent "visual event sequences" through lightweight spatio-temporal feature extraction, adaptive temporal pooling, and conceptual quantization with an event coherence bias. These visual event sequences are then fed into an LVLM-based Action Reasoning module, specifically a frozen LLaVA-1.5 model, adapted using parameter-efficient Prompt Tuning (P-Tuning v2) for action classification. Comprehensive evaluations on the NTU RGB+D and NTU RGB+D 120 datasets demonstrate that VT-LVLM-AR consistently achieves state-of-the-art performance, surpassing existing methods (e.g., 94.1% accuracy on NTU RGB+D X-Sub). Ablation studies confirm the critical contributions of VTEM's components and the efficacy of Prompt Tuning, while human evaluations underscore the interpretability of our visual event representations. This work highlights the immense potential of leveraging LVLMs for robust and interpretable video action understanding through effective video-to-language translation and efficient model adaptation.

Video Diffusion Models (VDMs) offer a promising approach for simulating dynamic scenes and environments, with broad applications in robotics and media generation. However, existing models often generate temporally incoherent content that violates basic physical intuition, significantly limiting their practical applicability. We propose PhysAlign, an efficient framework for physics-coherent image-to-video (I2V) generation that explicitly addresses this limitation. To overcome the critical scarcity of physics-annotated videos, we first construct a fully controllable synthetic data generation pipeline based on rigid-body simulation, yielding a highly-curated dataset with accurate, fine-grained physics and 3D annotations. Leveraging this data, PhysAlign constructs a unified physical latent space by coupling explicit 3D geometry constraints with a Gram-based spatio-temporal relational alignment that extracts kinematic priors from video foundation models. Extensive experiments demonstrate that PhysAlign significantly outperforms existing VDMs on tasks requiring complex physical reasoning and temporal stability, without compromising zero-shot visual quality. PhysAlign shows the potential to bridge the gap between raw visual synthesis and rigid-body kinematics, establishing a practical paradigm for genuinely physics-grounded video generation. The project page is available at https://physalign.github.io/PhysAlign.

We present Recurrent Video Masked-Autoencoders (RVM): a novel approach to video representation learning that leverages recurrent computation to model the temporal structure of video data. RVM couples an asymmetric masking objective with a transformer-based recurrent neural network to aggregate information over time, training solely on a simple pixel reconstruction loss. This design yields a highly efficient "generalist" encoder: RVM achieves competitive performance with state-of-the-art video models (e.g. VideoMAE, V-JEPA) on video-level tasks like action classification, and point and object tracking, while matching or exceeding the performance of image models (e.g. DINOv2) on tasks that require strong geometric and dense spatial features. Notably, RVM achieves strong performance in the small-model regime without requiring knowledge distillation, exhibiting up to 30x greater parameter efficiency than competing video masked autoencoders. Finally, we demonstrate that RVM's recurrent nature allows for stable feature propagation over long temporal horizons with linear computational cost, overcoming some of the limitations of standard spatio-temporal attention-based video models. Ablation studies further highlight the factors driving the model's success, with qualitative results showing that RVM learns rich representations of scene semantics, structure, and motion.

Video inpainting aims to fill spatio-temporal "corrupted" regions with plausible content. To achieve this goal, it is necessary to find correspondences from neighbouring frames to faithfully hallucinate the unknown content. Current methods achieve this goal through attention, flow-based warping, or 3D temporal convolution. However, flow-based warping can create artifacts when optical flow is not accurate, while temporal convolution may suffer from spatial misalignment. We propose 'Progressive Temporal Feature Alignment Network', which progressively enriches features extracted from the current frame with the feature warped from neighbouring frames using optical flow. Our approach corrects the spatial misalignment in the temporal feature propagation stage, greatly improving visual quality and temporal consistency of the inpainted videos. Using the proposed architecture, we achieve state-of-the-art performance on the DAVIS and FVI datasets compared to existing deep learning approaches. Code is available at https://github.com/MaureenZOU/TSAM.

Image generative models have made significant progress in generating realistic and diverse images, supported by comprehensive guidance from various evaluation metrics. However, current video generative models struggle to generate even short video clips, with limited tools that provide insights for improvements. Current video evaluation metrics are simple adaptations of image metrics by switching the embeddings with video embedding networks, which may underestimate the unique characteristics of video. Our analysis reveals that the widely used Frechet Video Distance (FVD) has a stronger emphasis on the spatial aspect than the temporal naturalness of video and is inherently constrained by the input size of the embedding networks used, limiting it to 16 frames. Additionally, it demonstrates considerable instability and diverges from human evaluations. To address the limitations, we propose STREAM, a new video evaluation metric uniquely designed to independently evaluate spatial and temporal aspects. This feature allows comprehensive analysis and evaluation of video generative models from various perspectives, unconstrained by video length. We provide analytical and experimental evidence demonstrating that STREAM provides an effective evaluation tool for both visual and temporal quality of videos, offering insights into area of improvement for video generative models. To the best of our knowledge, STREAM is the first evaluation metric that can separately assess the temporal and spatial aspects of videos. Our code is available at https://github.com/pro2nit/STREAM.

Surgical instrument segmentation under Federated Learning (FL) is a promising direction, which enables multiple surgical sites to collaboratively train the model without centralizing datasets. However, there exist very limited FL works in surgical data science, and FL methods for other modalities do not consider inherent characteristics in surgical domain: i) different scenarios show diverse anatomical backgrounds while highly similar instrument representation; ii) there exist surgical simulators which promote large-scale synthetic data generation with minimal efforts. In this paper, we propose a novel Personalized FL scheme, Spatio-Temporal Representation Decoupling and Enhancement (FedST), which wisely leverages surgical domain knowledge during both local-site and global-server training to boost segmentation. Concretely, our model embraces a Representation Separation and Cooperation (RSC) mechanism in local-site training, which decouples the query embedding layer to be trained privately, to encode respective backgrounds. Meanwhile, other parameters are optimized globally to capture the consistent representations of instruments, including the temporal layer to capture similar motion patterns. A textual-guided channel selection is further designed to highlight site-specific features, facilitating model adapta tion to each site. Moreover, in global-server training, we propose Synthesis-based Explicit Representation Quantification (SERQ), which defines an explicit representation target based on synthetic data to synchronize the model convergence during fusion for improving model generalization.

We introduce a novel FSVOS model that employs a local matching strategy to restrict the search space to the most relevant neighboring pixels. Rather than relying on inefficient standard im2col-like implementations (e.g., spatial convolutions, depthwise convolutions and feature-shifting mechanisms) or hardware-specific CUDA kernels (e.g., deformable and neighborhood attention), which often suffer from limited portability across non-CUDA devices, we reorganize the local sampling process through a direction-based sampling perspective. Specifically, we implement a non-parametric sampling mechanism that enables dynamically varying sampling regions. This approach provides the flexibility to adapt to diverse spatial structures without the computational costs of parametric layers and the need for model retraining. To further enhance feature coherence across frames, we design a supervised spatio-temporal contrastive learning scheme that enforces consistency in feature representations. In addition, we introduce a publicly available benchmark dataset for multi-object segmentation in X-ray angiography videos (MOSXAV), featuring detailed, manually labeled segmentation ground truth. Extensive experiments on the CADICA, XACV, and MOSXAV datasets show that our proposed FSVOS method outperforms current state-of-the-art video segmentation methods in terms of segmentation accuracy and generalization capability (i.e., seen and unseen categories). This work offers enhanced flexibility and potential for a wide range of clinical applications.

3D convolution neural networks (CNNs) have been the prevailing option for video recognition. To capture the temporal information, 3D convolutions are computed along the sequences, leading to cubically growing and expensive computations. To reduce the computational cost, previous methods resort to manually designed 3D/2D CNN structures with approximations or automatic search, which sacrifice the modeling ability or make training time-consuming. In this work, we propose to automatically design efficient 3D CNN architectures via a novel training-free neural architecture search approach tailored for 3D CNNs considering the model complexity. To measure the expressiveness of 3D CNNs efficiently, we formulate a 3D CNN as an information system and derive an analytic entropy score, based on the Maximum Entropy Principle. Specifically, we propose a spatio-temporal entropy score (STEntr-Score) with a refinement factor to handle the discrepancy of visual information in spatial and temporal dimensions, through dynamically leveraging the correlation between the feature map size and kernel size depth-wisely. Highly efficient and expressive 3D CNN architectures, \ie entropy-based 3D CNNs (E3D family), can then be efficiently searched by maximizing the STEntr-Score under a given computational budget, via an evolutionary algorithm without training the network parameters. Extensive experiments on Something-Something V1\&V2 and Kinetics400 demonstrate that the E3D family achieves state-of-the-art performance with higher computational efficiency. Code is available at https://github.com/alibaba/lightweight-neural-architecture-search.

Spatio-temporal scene graphs provide a principled representation for modeling evolving object interactions, yet existing methods remain fundamentally frame-centric: they reason only about currently visible objects, discard entities upon occlusion, and operate in 2D. To address this, we first introduce ActionGenome4D, a dataset that upgrades Action Genome videos into 4D scenes via feed-forward 3D reconstruction, world-frame oriented bounding boxes for every object involved in actions, and dense relationship annotations including for objects that are temporarily unobserved due to occlusion or camera motion. Building on this data, we formalize World Scene Graph Generation (WSGG), the task of constructing a world scene graph at each timestamp that encompasses all interacting objects in the scene, both observed and unobserved. We then propose three complementary methods, each exploring a different inductive bias for reasoning about unobserved objects: PWG (Persistent World Graph), which implements object permanence via a zero-order feature buffer; MWAE (Masked World Auto-Encoder), which reframes unobserved-object reasoning as masked completion with cross-view associative retrieval; and 4DST (4D Scene Transformer), which replaces the static buffer with differentiable per-object temporal attention enriched by 3D motion and camera-pose features. We further design and evaluate the performance of strong open-source Vision-Language Models on the WSGG task via a suite of Graph RAG-based approaches, establishing baselines for unlocalized relationship prediction. WSGG thus advances video scene understanding toward world-centric, temporally persistent, and interpretable scene reasoning.

Adapting pre-trained image models to video modality has proven to be an effective strategy for robust few-shot action recognition. In this work, we explore the potential of adapter tuning in image-to-video model adaptation and propose a novel video adapter tuning framework, called Disentangled-and-Deformable Spatio-Temporal Adapter (D^2ST-Adapter). It features a lightweight design, low adaptation overhead and powerful spatio-temporal feature adaptation capabilities. D^2ST-Adapter is structured with an internal dual-pathway architecture that enables built-in disentangled encoding of spatial and temporal features within the adapter, seamlessly integrating into the single-stream feature learning framework of pre-trained image models. In particular, we develop an efficient yet effective implementation of the D^2ST-Adapter, incorporating the specially devised anisotropic Deformable Spatio-Temporal Attention as its pivotal operation. This mechanism can be individually tailored for two pathways with anisotropic sampling densities along the spatial and temporal domains in 3D spatio-temporal space, enabling disentangled encoding of spatial and temporal features while maintaining a lightweight design. Extensive experiments by instantiating our method on both pre-trained ResNet and ViT demonstrate the superiority of our method over state-of-the-art methods. Our method is particularly well-suited to challenging scenarios where temporal dynamics are critical for action recognition. Code is available at https://github.com/qizhongtan/D2ST-Adapter.

Precise Event Spotting (PES) aims to identify events and their class from long, untrimmed videos, particularly in sports. The main objective of PES is to detect the event at the exact moment it occurs. Existing methods mainly rely on features from a large pre-trained network, which may not be ideal for the task. Furthermore, these methods overlook the issue of imbalanced event class distribution present in the data, negatively impacting performance in challenging scenarios. This paper demonstrates that an appropriately designed network, trained end-to-end, can outperform state-of-the-art (SOTA) methods. Particularly, we propose a network with a convolutional spatial-temporal feature extractor enhanced with our proposed Adaptive Spatio-Temporal Refinement Module (ASTRM) and a long-range temporal module. The ASTRM enhances the features with spatio-temporal information. Meanwhile, the long-range temporal module helps extract global context from the data by modeling long-range dependencies. To address the class imbalance issue, we introduce the Soft Instance Contrastive (SoftIC) loss that promotes feature compactness and class separation. Extensive experiments show that the proposed method is efficient and outperforms the SOTA methods, specifically in more challenging settings.

We present an efficient encoder-free approach for video-language understanding that achieves competitive performance while significantly reducing computational overhead. Current video-language models typically rely on heavyweight image encoders (300M-1.1B parameters) or video encoders (1B-1.4B parameters), creating a substantial computational burden when processing multi-frame videos. Our method introduces a novel Spatio-Temporal Alignment Block (STAB) that directly processes video inputs without requiring pre-trained encoders while using only 45M parameters for visual processing - at least a 6.5times reduction compared to traditional approaches. The STAB architecture combines Local Spatio-Temporal Encoding for fine-grained feature extraction, efficient spatial downsampling through learned attention and separate mechanisms for modeling frame-level and video-level relationships. Our model achieves comparable or superior performance to encoder-based approaches for open-ended video question answering on standard benchmarks. The fine-grained video question-answering evaluation demonstrates our model's effectiveness, outperforming the encoder-based approaches Video-ChatGPT and Video-LLaVA in key aspects like correctness and temporal understanding. Extensive ablation studies validate our architectural choices and demonstrate the effectiveness of our spatio-temporal modeling approach while achieving 3-4times faster processing speeds than previous methods. Code is available at https://github.com/jh-yi/Video-Panda.

Personalizing generative text-to-image models has seen remarkable progress, but extending this personalization to text-to-video models presents unique challenges. Unlike static concepts, personalizing text-to-video models has the potential to capture dynamic concepts, i.e., entities defined not only by their appearance but also by their motion. In this paper, we introduce Set-and-Sequence, a novel framework for personalizing Diffusion Transformers (DiTs)-based generative video models with dynamic concepts. Our approach imposes a spatio-temporal weight space within an architecture that does not explicitly separate spatial and temporal features. This is achieved in two key stages. First, we fine-tune Low-Rank Adaptation (LoRA) layers using an unordered set of frames from the video to learn an identity LoRA basis that represents the appearance, free from temporal interference. In the second stage, with the identity LoRAs frozen, we augment their coefficients with Motion Residuals and fine-tune them on the full video sequence, capturing motion dynamics. Our Set-and-Sequence framework results in a spatio-temporal weight space that effectively embeds dynamic concepts into the video model's output domain, enabling unprecedented editability and compositionality while setting a new benchmark for personalizing dynamic concepts.

We present Omni-RGPT, a multimodal large language model designed to facilitate region-level comprehension for both images and videos. To achieve consistent region representation across spatio-temporal dimensions, we introduce Token Mark, a set of tokens highlighting the target regions within the visual feature space. These tokens are directly embedded into spatial regions using region prompts (e.g., boxes or masks) and simultaneously incorporated into the text prompt to specify the target, establishing a direct connection between visual and text tokens. To further support robust video understanding without requiring tracklets, we introduce an auxiliary task that guides Token Mark by leveraging the consistency of the tokens, enabling stable region interpretation across the video. Additionally, we introduce a large-scale region-level video instruction dataset (RegVID-300k). Omni-RGPT achieves state-of-the-art results on image and video-based commonsense reasoning benchmarks while showing strong performance in captioning and referring expression comprehension tasks.

Diffusion-based video editing has emerged as an important paradigm for high-quality and flexible content generation. However, despite their generality and strong modeling capacity, Diffusion Transformers (DiT) remain computationally expensive due to the iterative denoising process, posing challenges for practical deployment. Existing video diffusion acceleration methods primarily exploit denoising timestep-level feature reuse, which mitigates the redundancy in denoising process, but overlooks the architectural redundancy within the DiT that many attention operations over spatio-temporal tokens are redundantly executed, offering little to no incremental contribution to the model output. This work introduces HetCache, a training-free diffusion acceleration framework designed to exploit the inherent heterogeneity in diffusion-based masked video-to-video (MV2V) generation and editing. Instead of uniformly reuse or randomly sampling tokens, HetCache assesses the contextual relevance and interaction strength among various types of tokens in designated computing steps. Guided by spatial priors, it divides the spatial-temporal tokens in DiT model into context and generative tokens, and selectively caches the context tokens that exhibit the strongest correlation and most representative semantics with generative ones. This strategy reduces redundant attention operations while maintaining editing consistency and fidelity. Experiments show that HetCache achieves a noticeable acceleration, including a 2.67times latency speedup and FLOPs reduction over commonly used foundation models, with negligible degradation in editing quality.

Egocentric video action recognition under domain shifts remains challenging due to large intra-class spatio-temporal variability, long-tailed feature distributions, and strong correlations between actions and environments. Existing benchmarks for egocentric domain generalization often conflate covariate shifts with concept shifts, making it difficult to reliably evaluate a model's ability to generalize across input distributions. To address this limitation, we introduce Ego4OOD, a domain generalization benchmark derived from Ego4D that emphasizes measurable covariate diversity while reducing concept shift through semantically coherent, moment-level action categories. Ego4OOD spans eight geographically distinct domains and is accompanied by a clustering-based covariate shift metric that provides a quantitative proxy for domain difficulty. We further leverage a one-vs-all binary training objective that decomposes multi-class action recognition into independent binary classification tasks. This formulation is particularly well-suited for covariate shift by reducing interference between visually similar classes under feature distribution shift. Using this formulation, we show that a lightweight two-layer fully connected network achieves performance competitive with state-of-the-art egocentric domain generalization methods on both Argo1M and Ego4OOD, despite using fewer parameters and no additional modalities. Our empirical analysis demonstrates a clear relationship between measured covariate shift and recognition performance, highlighting the importance of controlled benchmarks and quantitative domain characterization for studying out-of-distribution generalization in egocentric video.

Video super-resolution (VSR) aims to restore a sequence of high-resolution (HR) frames from their low-resolution (LR) counterparts. Although some progress has been made, there are grand challenges to effectively utilize temporal dependency in entire video sequences. Existing approaches usually align and aggregate video frames from limited adjacent frames (e.g., 5 or 7 frames), which prevents these approaches from satisfactory results. In this paper, we take one step further to enable effective spatio-temporal learning in videos. We propose a novel Trajectory-aware Transformer for Video Super-Resolution (TTVSR). In particular, we formulate video frames into several pre-aligned trajectories which consist of continuous visual tokens. For a query token, self-attention is only learned on relevant visual tokens along spatio-temporal trajectories. Compared with vanilla vision Transformers, such a design significantly reduces the computational cost and enables Transformers to model long-range features. We further propose a cross-scale feature tokenization module to overcome scale-changing problems that often occur in long-range videos. Experimental results demonstrate the superiority of the proposed TTVSR over state-of-the-art models, by extensive quantitative and qualitative evaluations in four widely-used video super-resolution benchmarks. Both code and pre-trained models can be downloaded at https://github.com/researchmm/TTVSR.

Large numbers of synthesized videos from diffusion models pose threats to information security and authenticity, leading to an increasing demand for generated content detection. However, existing video-level detection algorithms primarily focus on detecting facial forgeries and often fail to identify diffusion-generated content with a diverse range of semantics. To advance the field of video forensics, we propose an innovative algorithm named Multi-Modal Detection(MM-Det) for detecting diffusion-generated videos. MM-Det utilizes the profound perceptual and comprehensive abilities of Large Multi-modal Models (LMMs) by generating a Multi-Modal Forgery Representation (MMFR) from LMM's multi-modal space, enhancing its ability to detect unseen forgery content. Besides, MM-Det leverages an In-and-Across Frame Attention (IAFA) mechanism for feature augmentation in the spatio-temporal domain. A dynamic fusion strategy helps refine forgery representations for the fusion. Moreover, we construct a comprehensive diffusion video dataset, called Diffusion Video Forensics (DVF), across a wide range of forgery videos. MM-Det achieves state-of-the-art performance in DVF, demonstrating the effectiveness of our algorithm. Both source code and DVF are available at https://github.com/SparkleXFantasy/MM-Det.

Camera-based 3D object detection in BEV (Bird's Eye View) space has drawn great attention over the past few years. Dense detectors typically follow a two-stage pipeline by first constructing a dense BEV feature and then performing object detection in BEV space, which suffers from complex view transformations and high computation cost. On the other side, sparse detectors follow a query-based paradigm without explicit dense BEV feature construction, but achieve worse performance than the dense counterparts. In this paper, we find that the key to mitigate this performance gap is the adaptability of the detector in both BEV and image space. To achieve this goal, we propose SparseBEV, a fully sparse 3D object detector that outperforms the dense counterparts. SparseBEV contains three key designs, which are (1) scale-adaptive self attention to aggregate features with adaptive receptive field in BEV space, (2) adaptive spatio-temporal sampling to generate sampling locations under the guidance of queries, and (3) adaptive mixing to decode the sampled features with dynamic weights from the queries. On the test split of nuScenes, SparseBEV achieves the state-of-the-art performance of 67.5 NDS. On the val split, SparseBEV achieves 55.8 NDS while maintaining a real-time inference speed of 23.5 FPS. Code is available at https://github.com/MCG-NJU/SparseBEV.

Despite the fact real-world video deinterlacing and demosaicing are well-suited to supervised learning from synthetically degraded data because the degradation models are known and fixed, learned video deinterlacing and demosaicing have received much less attention compared to denoising and super-resolution tasks. We propose a new multi-picture architecture for video deinterlacing or demosaicing by aligning multiple supporting pictures with missing data to a reference picture to be reconstructed, benefiting from both local and global spatio-temporal correlations in the feature space using modified deformable convolution blocks and a novel residual efficient top-k self-attention (kSA) block, respectively. Separate reconstruction blocks are used to estimate different types of missing data. Our extensive experimental results, on synthetic or real-world datasets, demonstrate that the proposed novel architecture provides superior results that significantly exceed the state-of-the-art for both tasks in terms of PSNR, SSIM, and perceptual quality. Ablation studies are provided to justify and show the benefit of each novel modification made to the deformable convolution and residual efficient kSA blocks. Code is available: https://github.com/KUIS-AI-Tekalp-Research-Group/Video-Deinterlacing.

In this paper, we present BEVerse, a unified framework for 3D perception and prediction based on multi-camera systems. Unlike existing studies focusing on the improvement of single-task approaches, BEVerse features in producing spatio-temporal Birds-Eye-View (BEV) representations from multi-camera videos and jointly reasoning about multiple tasks for vision-centric autonomous driving. Specifically, BEVerse first performs shared feature extraction and lifting to generate 4D BEV representations from multi-timestamp and multi-view images. After the ego-motion alignment, the spatio-temporal encoder is utilized for further feature extraction in BEV. Finally, multiple task decoders are attached for joint reasoning and prediction. Within the decoders, we propose the grid sampler to generate BEV features with different ranges and granularities for different tasks. Also, we design the method of iterative flow for memory-efficient future prediction. We show that the temporal information improves 3D object detection and semantic map construction, while the multi-task learning can implicitly benefit motion prediction. With extensive experiments on the nuScenes dataset, we show that the multi-task BEVerse outperforms existing single-task methods on 3D object detection, semantic map construction, and motion prediction. Compared with the sequential paradigm, BEVerse also favors in significantly improved efficiency. The code and trained models will be released at https://github.com/zhangyp15/BEVerse.

Multi-view generation with camera pose control and prompt-based customization are both essential elements for achieving controllable generative models. However, existing multi-view generation models do not support customization with geometric consistency, whereas customization models lack explicit viewpoint control, making them challenging to unify. Motivated by these gaps, we introduce a novel task, multi-view customization, which aims to jointly achieve multi-view camera pose control and customization. Due to the scarcity of training data in customization, existing multi-view generation models, which inherently rely on large-scale datasets, struggle to generalize to diverse prompts. To address this, we propose MVCustom, a novel diffusion-based framework explicitly designed to achieve both multi-view consistency and customization fidelity. In the training stage, MVCustom learns the subject's identity and geometry using a feature-field representation, incorporating the text-to-video diffusion backbone enhanced with dense spatio-temporal attention, which leverages temporal coherence for multi-view consistency. In the inference stage, we introduce two novel techniques: depth-aware feature rendering explicitly enforces geometric consistency, and consistent-aware latent completion ensures accurate perspective alignment of the customized subject and surrounding backgrounds. Extensive experiments demonstrate that MVCustom is the only framework that simultaneously achieves faithful multi-view generation and customization.

Traditional rendering pipelines rely on complex assets, accurate materials and lighting, and substantial computational resources to produce realistic imagery, yet they still face challenges in scalability and realism for populated dynamic scenes. We present C2R (Coarse-to-Real), a generative rendering framework that synthesizes real-style urban crowd videos from coarse 3D simulations. Our approach uses coarse 3D renderings to explicitly control scene layout, camera motion, and human trajectories, while a learned neural renderer generates realistic appearance, lighting, and fine-scale dynamics guided by text prompts. To overcome the lack of paired training data between coarse simulations and real videos, we adopt a two-stage synthetic-real domain-hedging strategy that first learns a strong generative prior from large-scale real footage, and then introduces controllability by using a small amount of paired synthetic coarse-to-fine data to anchor shared implicit spatio-temporal features across domains. The resulting system supports coarse-to-fine control, generalizes across diverse CG and game inputs, and produces temporally consistent, controllable, and realistic urban scene videos from minimal 3D input. We will release the model and project webpage at https://gonzalognogales.github.io/coarse2real/.

The paucity of videos in current action classification datasets (UCF-101 and HMDB-51) has made it difficult to identify good video architectures, as most methods obtain similar performance on existing small-scale benchmarks. This paper re-evaluates state-of-the-art architectures in light of the new Kinetics Human Action Video dataset. Kinetics has two orders of magnitude more data, with 400 human action classes and over 400 clips per class, and is collected from realistic, challenging YouTube videos. We provide an analysis on how current architectures fare on the task of action classification on this dataset and how much performance improves on the smaller benchmark datasets after pre-training on Kinetics. We also introduce a new Two-Stream Inflated 3D ConvNet (I3D) that is based on 2D ConvNet inflation: filters and pooling kernels of very deep image classification ConvNets are expanded into 3D, making it possible to learn seamless spatio-temporal feature extractors from video while leveraging successful ImageNet architecture designs and even their parameters. We show that, after pre-training on Kinetics, I3D models considerably improve upon the state-of-the-art in action classification, reaching 80.9% on HMDB-51 and 98.0% on UCF-101.

Optical flow models trained on high-quality data often degrade severely when confronted with real-world corruptions such as blur, noise, and compression artifacts. To overcome this limitation, we formulate Degradation-Aware Optical Flow, a new task targeting accurate dense correspondence estimation from real-world corrupted videos. Our key insight is that the intermediate representations of image restoration diffusion models are inherently corruption-aware but lack temporal awareness. To address this limitation, we lift the model to attend across adjacent frames via full spatio-temporal attention, and empirically demonstrate that the resulting features exhibit zero-shot correspondence capabilities. Based on this finding, we present DA-Flow, a hybrid architecture that fuses these diffusion features with convolutional features within an iterative refinement framework. DA-Flow substantially outperforms existing optical flow methods under severe degradation across multiple benchmarks.

We present a large-scale study on unsupervised spatiotemporal representation learning from videos. With a unified perspective on four recent image-based frameworks, we study a simple objective that can easily generalize all these methods to space-time. Our objective encourages temporally-persistent features in the same video, and in spite of its simplicity, it works surprisingly well across: (i) different unsupervised frameworks, (ii) pre-training datasets, (iii) downstream datasets, and (iv) backbone architectures. We draw a series of intriguing observations from this study, e.g., we discover that encouraging long-spanned persistency can be effective even if the timespan is 60 seconds. In addition to state-of-the-art results in multiple benchmarks, we report a few promising cases in which unsupervised pre-training can outperform its supervised counterpart. Code is made available at https://github.com/facebookresearch/SlowFast

We introduce VideoMamba, a novel adaptation of the pure Mamba architecture, specifically designed for video recognition. Unlike transformers that rely on self-attention mechanisms leading to high computational costs by quadratic complexity, VideoMamba leverages Mamba's linear complexity and selective SSM mechanism for more efficient processing. The proposed Spatio-Temporal Forward and Backward SSM allows the model to effectively capture the complex relationship between non-sequential spatial and sequential temporal information in video. Consequently, VideoMamba is not only resource-efficient but also effective in capturing long-range dependency in videos, demonstrated by competitive performance and outstanding efficiency on a variety of video understanding benchmarks. Our work highlights the potential of VideoMamba as a powerful tool for video understanding, offering a simple yet effective baseline for future research in video analysis.

Recently, with the emergence of large language models, multimodal LLMs have demonstrated exceptional capabilities in image and video modalities. Despite advancements in video comprehension, the substantial computational demands of long video sequences lead current video LLMs (Vid-LLMs) to employ compression strategies at both the inter-frame level (e.g., uniform sampling of video frames) and intra-frame level (e.g., condensing all visual tokens of each frame into a limited number). However, this approach often neglects the uneven temporal distribution of critical information across frames, risking the omission of keyframes that contain essential temporal and semantic details. To tackle these challenges, we propose KFFocus, a method designed to efficiently compress video tokens and emphasize the informative context present within video frames. We substitute uniform sampling with a refined approach inspired by classic video compression principles to identify and capture keyframes based on their temporal redundancy. By assigning varying condensation ratios to frames based on their contextual relevance, KFFocus efficiently reduces token redundancy while preserving informative content details. Additionally, we introduce a spatiotemporal modeling module that encodes both the temporal relationships between video frames and the spatial structure within each frame, thus providing Vid-LLMs with a nuanced understanding of spatial-temporal dynamics. Extensive experiments on widely recognized video understanding benchmarks, especially long video scenarios, demonstrate that KFFocus significantly outperforms existing methods, achieving substantial computational efficiency and enhanced accuracy.

Although Video Large Language Models (VLLMs) have shown remarkable capabilities in video understanding, they are required to process high volumes of visual tokens, causing significant computational inefficiency. Existing VLLMs acceleration frameworks usually compress spatial and temporal redundancy independently, which overlooks the spatiotemporal relationships, thereby leading to suboptimal spatiotemporal compression. The highly correlated visual features are likely to change in spatial position, scale, orientation, and other attributes over time due to the dynamic nature of video. Building on this insight, we introduce FlashVID, a training-free inference acceleration framework for VLLMs. Specifically, FlashVID utilizes Attention and Diversity-based Token Selection (ADTS) to select the most representative tokens for basic video representation, then applies Tree-based Spatiotemporal Token Merging (TSTM) for fine-grained spatiotemporal redundancy elimination. Extensive experiments conducted on three representative VLLMs across five video understanding benchmarks demonstrate the effectiveness and generalization of our method. Notably, by retaining only 10% of visual tokens, FlashVID preserves 99.1% of the performance of LLaVA-OneVision. Consequently, FlashVID can serve as a training-free and plug-and-play module for extending long video frames, which enables a 10x increase in video frame input to Qwen2.5-VL, resulting in a relative improvement of 8.6% within the same computational budget. Code is available at https://github.com/Fanziyang-v/FlashVID.

Video Anomaly Detection~(VAD) focuses on identifying anomalies within videos. Supervised methods require an amount of in-domain training data and often struggle to generalize to unseen anomalies. In contrast, training-free methods leverage the intrinsic world knowledge of large language models (LLMs) to detect anomalies but face challenges in localizing fine-grained visual transitions and diverse events. Therefore, we propose EventVAD, an event-aware video anomaly detection framework that combines tailored dynamic graph architectures and multimodal LLMs through temporal-event reasoning. Specifically, EventVAD first employs dynamic spatiotemporal graph modeling with time-decay constraints to capture event-aware video features. Then, it performs adaptive noise filtering and uses signal ratio thresholding to detect event boundaries via unsupervised statistical features. The statistical boundary detection module reduces the complexity of processing long videos for MLLMs and improves their temporal reasoning through event consistency. Finally, it utilizes a hierarchical prompting strategy to guide MLLMs in performing reasoning before determining final decisions. We conducted extensive experiments on the UCF-Crime and XD-Violence datasets. The results demonstrate that EventVAD with a 7B MLLM achieves state-of-the-art (SOTA) in training-free settings, outperforming strong baselines that use 7B or larger MLLMs.

We study joint video and language (VL) pre-training to enable cross-modality learning and benefit plentiful downstream VL tasks. Existing works either extract low-quality video features or learn limited text embedding, while neglecting that high-resolution videos and diversified semantics can significantly improve cross-modality learning. In this paper, we propose a novel High-resolution and Diversified VIdeo-LAnguage pre-training model (HD-VILA) for many visual tasks. In particular, we collect a large dataset with two distinct properties: 1) the first high-resolution dataset including 371.5k hours of 720p videos, and 2) the most diversified dataset covering 15 popular YouTube categories. To enable VL pre-training, we jointly optimize the HD-VILA model by a hybrid Transformer that learns rich spatiotemporal features, and a multimodal Transformer that enforces interactions of the learned video features with diversified texts. Our pre-training model achieves new state-of-the-art results in 10 VL understanding tasks and 2 more novel text-to-visual generation tasks. For example, we outperform SOTA models with relative increases of 40.4% R@1 in zero-shot MSR-VTT text-to-video retrieval task and 55.4% in high-resolution dataset LSMDC. The learned VL embedding is also effective in generating visually pleasing and semantically relevant results in text-to-visual editing and super-resolution tasks.

We present the task of Spatio-Temporal Video Question Answering, which requires intelligent systems to simultaneously retrieve relevant moments and detect referenced visual concepts (people and objects) to answer natural language questions about videos. We first augment the TVQA dataset with 310.8K bounding boxes, linking depicted objects to visual concepts in questions and answers. We name this augmented version as TVQA+. We then propose Spatio-Temporal Answerer with Grounded Evidence (STAGE), a unified framework that grounds evidence in both spatial and temporal domains to answer questions about videos. Comprehensive experiments and analyses demonstrate the effectiveness of our framework and how the rich annotations in our TVQA+ dataset can contribute to the question answering task. Moreover, by performing this joint task, our model is able to produce insightful and interpretable spatio-temporal attention visualizations. Dataset and code are publicly available at: http: //tvqa.cs.unc.edu, https://github.com/jayleicn/TVQAplus

Pre-training on large-scale video data has become a common recipe for learning transferable spatiotemporal representations in recent years. Despite some progress, existing methods are mostly limited to highly curated datasets (e.g., K400) and exhibit unsatisfactory out-of-the-box representations. We argue that it is due to the fact that they only capture pixel-level knowledge rather than spatiotemporal semantics, which hinders further progress in video understanding. Inspired by the great success of image-text pre-training (e.g., CLIP), we take the first step to exploit language semantics to boost transferable spatiotemporal representation learning. We introduce a new pretext task, Turning to Video for Transcript Sorting (TVTS), which sorts shuffled ASR scripts by attending to learned video representations. We do not rely on descriptive captions and learn purely from video, i.e., leveraging the natural transcribed speech knowledge to provide noisy but useful semantics over time. Our method enforces the vision model to contextualize what is happening over time so that it can re-organize the narrative transcripts, and can seamlessly apply to large-scale uncurated video data in the real world. Our method demonstrates strong out-of-the-box spatiotemporal representations on diverse benchmarks, e.g., +13.6% gains over VideoMAE on SSV2 via linear probing. The code is available at https://github.com/TencentARC/TVTS.

Videos carry rich visual information including object description, action, interaction, etc., but the existing multimodal large language models (MLLMs) fell short in referential understanding scenarios such as video-based referring. In this paper, we present Artemis, an MLLM that pushes video-based referential understanding to a finer level. Given a video, Artemis receives a natural-language question with a bounding box in any video frame and describes the referred target in the entire video. The key to achieving this goal lies in extracting compact, target-specific video features, where we set a solid baseline by tracking and selecting spatiotemporal features from the video. We train Artemis on the newly established VideoRef45K dataset with 45K video-QA pairs and design a computationally efficient, three-stage training procedure. Results are promising both quantitatively and qualitatively. Additionally, we show that \model can be integrated with video grounding and text summarization tools to understand more complex scenarios. Code and data are available at https://github.com/qiujihao19/Artemis.

Spatio-temporal consistency is a critical research topic in video generation. A qualified generated video segment must ensure plot plausibility and coherence while maintaining visual consistency of objects and scenes across varying viewpoints. Prior research, especially in open-source projects, primarily focuses on either temporal or spatial consistency, or their basic combination, such as appending a description of a camera movement after a prompt without constraining the outcomes of this movement. However, camera movement may introduce new objects to the scene or eliminate existing ones, thereby overlaying and affecting the preceding narrative. Especially in videos with numerous camera movements, the interplay between multiple plots becomes increasingly complex. This paper introduces and examines integral spatio-temporal consistency, considering the synergy between plot progression and camera techniques, and the long-term impact of prior content on subsequent generation. Our research encompasses dataset construction through to the development of the model. Initially, we constructed a DropletVideo-10M dataset, which comprises 10 million videos featuring dynamic camera motion and object actions. Each video is annotated with an average caption of 206 words, detailing various camera movements and plot developments. Following this, we developed and trained the DropletVideo model, which excels in preserving spatio-temporal coherence during video generation. The DropletVideo dataset and model are accessible at https://dropletx.github.io.

This paper proposes a novel pretext task to address the self-supervised video representation learning problem. Specifically, given an unlabeled video clip, we compute a series of spatio-temporal statistical summaries, such as the spatial location and dominant direction of the largest motion, the spatial location and dominant color of the largest color diversity along the temporal axis, etc. Then a neural network is built and trained to yield the statistical summaries given the video frames as inputs. In order to alleviate the learning difficulty, we employ several spatial partitioning patterns to encode rough spatial locations instead of exact spatial Cartesian coordinates. Our approach is inspired by the observation that human visual system is sensitive to rapidly changing contents in the visual field, and only needs impressions about rough spatial locations to understand the visual contents. To validate the effectiveness of the proposed approach, we conduct extensive experiments with four 3D backbone networks, i.e., C3D, 3D-ResNet, R(2+1)D and S3D-G. The results show that our approach outperforms the existing approaches across these backbone networks on four downstream video analysis tasks including action recognition, video retrieval, dynamic scene recognition, and action similarity labeling. The source code is publicly available at: https://github.com/laura-wang/video_repres_sts.

Understanding the content of videos is one of the core techniques for developing various helpful applications in the real world, such as recognizing various human actions for surveillance systems or customer behavior analysis in an autonomous shop. However, understanding the content or story of the video still remains a challenging problem due to its sheer amount of data and temporal structure. In this paper, we propose a multi-channel neural network structure that adopts a two-stream network structure, which has been shown high performance in human action recognition field, and use it as a spatiotemporal video feature extractor for solving video question and answering task. We also adopt a squeeze-and-excitation structure to two-stream network structure for achieving a channel-wise attended spatiotemporal feature. For jointly modeling the spatiotemporal features from video and the textual features from the question, we design a context matching module with a level adjusting layer to remove the gap of information between visual and textual features by applying attention mechanism on joint modeling. Finally, we adopt a scoring mechanism and smoothed ranking loss objective function for selecting the correct answer from answer candidates. We evaluate our model with TVQA dataset, and our approach shows the improved result in textual only setting, but the result with visual feature shows the limitation and possibility of our approach.

Autoregressive models for video generation typically operate frame-by-frame, extending next-token prediction from language to video's temporal dimension. We question that unlike word as token is universally agreed in language if frame is a appropriate prediction unit? To address this, we present VideoAR, a unified framework that supports a spectrum of prediction units including full frames, key-detail frames, multiscale refinements, and spatiotemporal cubes. Among these designs, we find model video generation using spatiotemporal cubes as prediction units, which allows autoregressive models to operate across both spatial and temporal dimensions simultaneously. This approach eliminates the assumption that frames are the natural atomic units for video autoregression. We evaluate VideoAR across diverse prediction strategies, finding that cube-based prediction consistently delivers superior quality, speed, and temporal coherence. By removing the frame-by-frame constraint, our video generator surpasses state-of-the-art baselines on VBench while achieving faster inference and enabling seamless scaling to minute-long sequences. We hope this work will motivate rethinking sequence decomposition in video and other spatiotemporal domains.

Recent progress in using recurrent neural networks (RNNs) for image description has motivated the exploration of their application for video description. However, while images are static, working with videos requires modeling their dynamic temporal structure and then properly integrating that information into a natural language description. In this context, we propose an approach that successfully takes into account both the local and global temporal structure of videos to produce descriptions. First, our approach incorporates a spatial temporal 3-D convolutional neural network (3-D CNN) representation of the short temporal dynamics. The 3-D CNN representation is trained on video action recognition tasks, so as to produce a representation that is tuned to human motion and behavior. Second we propose a temporal attention mechanism that allows to go beyond local temporal modeling and learns to automatically select the most relevant temporal segments given the text-generating RNN. Our approach exceeds the current state-of-art for both BLEU and METEOR metrics on the Youtube2Text dataset. We also present results on a new, larger and more challenging dataset of paired video and natural language descriptions.

Continuous space-time video super-resolution (C-STVSR) endeavors to upscale videos simultaneously at arbitrary spatial and temporal scales, which has recently garnered increasing interest. However, prevailing methods struggle to yield satisfactory videos at out-of-distribution spatial and temporal scales. On the other hand, event streams characterized by high temporal resolution and high dynamic range, exhibit compelling promise in vision tasks. This paper presents EvEnhancer, an innovative approach that marries the unique advantages of event streams to elevate effectiveness, efficiency, and generalizability for C-STVSR. Our approach hinges on two pivotal components: 1) Event-adapted synthesis capitalizes on the spatiotemporal correlations between frames and events to discern and learn long-term motion trajectories, enabling the adaptive interpolation and fusion of informative spatiotemporal features; 2) Local implicit video transformer integrates local implicit video neural function with cross-scale spatiotemporal attention to learn continuous video representations utilized to generate plausible videos at arbitrary resolutions and frame rates. Experiments show that EvEnhancer achieves superiority on synthetic and real-world datasets and preferable generalizability on out-of-distribution scales against state-of-the-art methods. Code is available at https://github.com/W-Shuoyan/EvEnhancer.

Significant progress has been made in spatial intelligence, spanning both spatial reconstruction and world exploration. However, the scalability and real-world fidelity of current models remain severely constrained by the scarcity of large-scale, high-quality training data. While several datasets provide camera pose information, they are typically limited in scale, diversity, and annotation richness, particularly for real-world dynamic scenes with ground-truth camera motion. To this end, we collect SpatialVID, a dataset consists of a large corpus of in-the-wild videos with diverse scenes, camera movements and dense 3D annotations such as per-frame camera poses, depth, and motion instructions. Specifically, we collect more than 21,000 hours of raw video, and process them into 2.7 million clips through a hierarchical filtering pipeline, totaling 7,089 hours of dynamic content. A subsequent annotation pipeline enriches these clips with detailed spatial and semantic information, including camera poses, depth maps, dynamic masks, structured captions, and serialized motion instructions. Analysis of SpatialVID's data statistics reveals a richness and diversity that directly foster improved model generalization and performance, establishing it as a key asset for the video and 3D vision research community.

Multimodal large language models (MLLMs) have made remarkable progress in either temporal or spatial localization. However, they struggle to perform spatio-temporal video grounding. This limitation stems from two major challenges. Firstly, it is difficult to extract accurate spatio-temporal information of each frame in the video. Secondly, the substantial number of visual tokens makes it challenging to precisely map visual tokens of each frame to their corresponding spatial coordinates. To address these issues, we introduce SpaceVLLM, a MLLM endowed with spatio-temporal video grounding capability. Specifically, we adopt a set of interleaved Spatio-Temporal Aware Queries to capture temporal perception and dynamic spatial information. Moreover, we propose a Query-Guided Space Decoder to establish a corresponding connection between the queries and spatial coordinates. Additionally, due to the lack of spatio-temporal datasets, we construct the Unified Spatio-Temporal Grounding (Uni-STG) dataset, comprising 480K instances across three tasks. This dataset fully exploits the potential of MLLM to simultaneously facilitate localization in both temporal and spatial dimensions. Extensive experiments demonstrate that SpaceVLLM achieves the state-of-the-art performance across 11 benchmarks covering temporal, spatial, spatio-temporal and video understanding tasks, highlighting the effectiveness of our approach. Our code, datasets and model will be released at https://github.com/Jayce1kk/SpaceVLLM.

We propose a self-supervised method to learn feature representations from videos. A standard approach in traditional self-supervised methods uses positive-negative data pairs to train with contrastive learning strategy. In such a case, different modalities of the same video are treated as positives and video clips from a different video are treated as negatives. Because the spatio-temporal information is important for video representation, we extend the negative samples by introducing intra-negative samples, which are transformed from the same anchor video by breaking temporal relations in video clips. With the proposed Inter-Intra Contrastive (IIC) framework, we can train spatio-temporal convolutional networks to learn video representations. There are many flexible options in our IIC framework and we conduct experiments by using several different configurations. Evaluations are conducted on video retrieval and video recognition tasks using the learned video representation. Our proposed IIC outperforms current state-of-the-art results by a large margin, such as 16.7% and 9.5% points improvements in top-1 accuracy on UCF101 and HMDB51 datasets for video retrieval, respectively. For video recognition, improvements can also be obtained on these two benchmark datasets. Code is available at https://github.com/BestJuly/Inter-intra-video-contrastive-learning.

The explosive growth of videos on streaming media platforms has underscored the urgent need for effective video quality assessment (VQA) algorithms to monitor and perceptually optimize the quality of streaming videos. However, VQA remains an extremely challenging task due to the diverse video content and the complex spatial and temporal distortions, thus necessitating more advanced methods to address these issues. Nowadays, large multimodal models (LMMs), such as GPT-4V, have exhibited strong capabilities for various visual understanding tasks, motivating us to leverage the powerful multimodal representation ability of LMMs to solve the VQA task. Therefore, we propose the first Large Multi-Modal Video Quality Assessment (LMM-VQA) model, which introduces a novel spatiotemporal visual modeling strategy for quality-aware feature extraction. Specifically, we first reformulate the quality regression problem into a question and answering (Q&A) task and construct Q&A prompts for VQA instruction tuning. Then, we design a spatiotemporal vision encoder to extract spatial and temporal features to represent the quality characteristics of videos, which are subsequently mapped into the language space by the spatiotemporal projector for modality alignment. Finally, the aligned visual tokens and the quality-inquired text tokens are aggregated as inputs for the large language model (LLM) to generate the quality score and level. Extensive experiments demonstrate that LMM-VQA achieves state-of-the-art performance across five VQA benchmarks, exhibiting an average improvement of 5% in generalization ability over existing methods. Furthermore, due to the advanced design of the spatiotemporal encoder and projector, LMM-VQA also performs exceptionally well on general video understanding tasks, further validating its effectiveness. Our code will be released at https://github.com/Sueqk/LMM-VQA.

We present pure-transformer based models for video classification, drawing upon the recent success of such models in image classification. Our model extracts spatio-temporal tokens from the input video, which are then encoded by a series of transformer layers. In order to handle the long sequences of tokens encountered in video, we propose several, efficient variants of our model which factorise the spatial- and temporal-dimensions of the input. Although transformer-based models are known to only be effective when large training datasets are available, we show how we can effectively regularise the model during training and leverage pretrained image models to be able to train on comparatively small datasets. We conduct thorough ablation studies, and achieve state-of-the-art results on multiple video classification benchmarks including Kinetics 400 and 600, Epic Kitchens, Something-Something v2 and Moments in Time, outperforming prior methods based on deep 3D convolutional networks. To facilitate further research, we release code at https://github.com/google-research/scenic/tree/main/scenic/projects/vivit

Effective spatio-temporal representation is fundamental to modeling, understanding, and predicting dynamics in videos. The atomic unit of a video, the pixel, traces a continuous 3D trajectory over time, serving as the primitive element of dynamics. Based on this principle, we propose representing any video as a Trajectory Field: a dense mapping that assigns a continuous 3D trajectory function of time to each pixel in every frame. With this representation, we introduce Trace Anything, a neural network that predicts the entire trajectory field in a single feed-forward pass. Specifically, for each pixel in each frame, our model predicts a set of control points that parameterizes a trajectory (i.e., a B-spline), yielding its 3D position at arbitrary query time instants. We trained the Trace Anything model on large-scale 4D data, including data from our new platform, and our experiments demonstrate that: (i) Trace Anything achieves state-of-the-art performance on our new benchmark for trajectory field estimation and performs competitively on established point-tracking benchmarks; (ii) it offers significant efficiency gains thanks to its one-pass paradigm, without requiring iterative optimization or auxiliary estimators; and (iii) it exhibits emergent abilities, including goal-conditioned manipulation, motion forecasting, and spatio-temporal fusion. Project page: https://trace-anything.github.io/.

Automatically recognizing surgical gestures is a crucial step towards a thorough understanding of surgical skill. Possible areas of application include automatic skill assessment, intra-operative monitoring of critical surgical steps, and semi-automation of surgical tasks. Solutions that rely only on the laparoscopic video and do not require additional sensor hardware are especially attractive as they can be implemented at low cost in many scenarios. However, surgical gesture recognition based only on video is a challenging problem that requires effective means to extract both visual and temporal information from the video. Previous approaches mainly rely on frame-wise feature extractors, either handcrafted or learned, which fail to capture the dynamics in surgical video. To address this issue, we propose to use a 3D Convolutional Neural Network (CNN) to learn spatiotemporal features from consecutive video frames. We evaluate our approach on recordings of robot-assisted suturing on a bench-top model, which are taken from the publicly available JIGSAWS dataset. Our approach achieves high frame-wise surgical gesture recognition accuracies of more than 84%, outperforming comparable models that either extract only spatial features or model spatial and low-level temporal information separately. For the first time, these results demonstrate the benefit of spatiotemporal CNNs for video-based surgical gesture recognition.

How to effectively explore spatial-temporal features is important for video colorization. Instead of stacking multiple frames along the temporal dimension or recurrently propagating estimated features that will accumulate errors or cannot explore information from far-apart frames, we develop a memory-based feature propagation module that can establish reliable connections with features from far-apart frames and alleviate the influence of inaccurately estimated features. To extract better features from each frame for the above-mentioned feature propagation, we explore the features from large-pretrained visual models to guide the feature estimation of each frame so that the estimated features can model complex scenarios. In addition, we note that adjacent frames usually contain similar contents. To explore this property for better spatial and temporal feature utilization, we develop a local attention module to aggregate the features from adjacent frames in a spatial-temporal neighborhood. We formulate our memory-based feature propagation module, large-pretrained visual model guided feature estimation module, and local attention module into an end-to-end trainable network (named ColorMNet) and show that it performs favorably against state-of-the-art methods on both the benchmark datasets and real-world scenarios. The source code and pre-trained models will be available at https://github.com/yyang181/colormnet.

Diffusion-based models have shown strong performance in video super-resolution (VSR) and video frame interpolation (VFI). However, their role in the coupled space-time video super-resolution (STVSR) setting remains limited. Existing diffusion-based STVSR approaches suffer from two issues: (1) low inference efficiency and (2) insufficient utilization of spatiotemporal information. These limitations impede deployment. To address these issues, we introduce DiffST, an efficient spatiotemporal-aware video diffusion framework for real-world STVSR. To improve efficiency, we adapt a pre-trained diffusion model for one-step sampling and process the entire video directly rather than operating on individual frames. Furthermore, to enhance spatiotemporal information utilization, we introduce cross-frame context aggregation (CFCA) and video representation guidance (VRG). The CFCA module aggregates information across multiple keyframes to produce intermediate frames. The VRG module extracts video-level global features to guide the diffusion process. Extensive experiments show that DiffST obtains leading results on real-world STVSR tasks. It also maintains high inference efficiency, running about 17times faster than previous diffusion-based STVSR methods. Code is available at: https://github.com/zhengchen1999/DiffST.

In this research, we present a novel approach to motion customization in video generation, addressing the widespread gap in the thorough exploration of motion representation within video generative models. Recognizing the unique challenges posed by video's spatiotemporal nature, our method introduces Motion Embeddings, a set of explicit, temporally coherent one-dimensional embeddings derived from a given video. These embeddings are designed to integrate seamlessly with the temporal transformer modules of video diffusion models, modulating self-attention computations across frames without compromising spatial integrity. Our approach offers a compact and efficient solution to motion representation and enables complex manipulations of motion characteristics through vector arithmetic in the embedding space. Furthermore, we identify the Temporal Discrepancy in video generative models, which refers to variations in how different motion modules process temporal relationships between frames. We leverage this understanding to optimize the integration of our motion embeddings. Our contributions include the introduction of a tailored motion embedding for customization tasks, insights into the temporal processing differences in video models, and a demonstration of the practical advantages and effectiveness of our method through extensive experiments.

We present VideoFactory, an innovative framework for generating high-quality open-domain videos. VideoFactory excels in producing high-definition (1376x768), widescreen (16:9) videos without watermarks, creating an engaging user experience. Generating videos guided by text instructions poses significant challenges, such as modeling the complex relationship between space and time, and the lack of large-scale text-video paired data. Previous approaches extend pretrained text-to-image generation models by adding temporal 1D convolution/attention modules for video generation. However, these approaches overlook the importance of jointly modeling space and time, inevitably leading to temporal distortions and misalignment between texts and videos. In this paper, we propose a novel approach that strengthens the interaction between spatial and temporal perceptions. In particular, we utilize a swapped cross-attention mechanism in 3D windows that alternates the "query" role between spatial and temporal blocks, enabling mutual reinforcement for each other. To fully unlock model capabilities for high-quality video generation, we curate a large-scale video dataset called HD-VG-130M. This dataset comprises 130 million text-video pairs from the open-domain, ensuring high-definition, widescreen and watermark-free characters. Objective metrics and user studies demonstrate the superiority of our approach in terms of per-frame quality, temporal correlation, and text-video alignment, with clear margins.

Spatio-temporal contexts are crucial in understanding human actions in videos. Recent state-of-the-art Convolutional Neural Network (ConvNet) based action recognition systems frequently involve 3D spatio-temporal ConvNet filters, chunking videos into fixed length clips and Long Short Term Memory (LSTM) networks. Such architectures are designed to take advantage of both short term and long term temporal contexts, but also requires the accumulation of a predefined number of video frames (e.g., to construct video clips for 3D ConvNet filters, to generate enough inputs for LSTMs). For applications that require low-latency online predictions of fast-changing action scenes, a new action recognition system is proposed in this paper. Termed "Weighted Multi-Region Convolutional Neural Network" (WMR ConvNet), the proposed system is LSTM-free, and is based on 2D ConvNet that does not require the accumulation of video frames for 3D ConvNet filtering. Unlike early 2D ConvNets that are based purely on RGB frames and optical flow frames, the WMR ConvNet is designed to simultaneously capture multiple spatial and short term temporal cues (e.g., human poses, occurrences of objects in the background) with both the primary region (foreground) and secondary regions (mostly background). On both the UCF101 and HMDB51 datasets, the proposed WMR ConvNet achieves the state-of-the-art performance among competing low-latency algorithms. Furthermore, WMR ConvNet even outperforms the 3D ConvNet based C3D algorithm that requires video frame accumulation. In an ablation study with the optical flow ConvNet stream removed, the ablated WMR ConvNet nevertheless outperforms competing algorithms.

The goal of spatial-temporal action detection is to determine the time and place where each person's action occurs in a video and classify the corresponding action category. Most of the existing methods adopt fully-supervised learning, which requires a large amount of training data, making it very difficult to achieve zero-shot learning. In this paper, we propose to utilize a pre-trained visual-language model to extract the representative image and text features, and model the relationship between these features through different interaction modules to obtain the interaction feature. In addition, we use this feature to prompt each label to obtain more appropriate text features. Finally, we calculate the similarity between the interaction feature and the text feature for each label to determine the action category. Our experiments on J-HMDB and UCF101-24 datasets demonstrate that the proposed interaction module and prompting make the visual-language features better aligned, thus achieving excellent accuracy for zero-shot spatio-temporal action detection. The code will be available at https://github.com/webber2933/iCLIP.

Long-context video understanding in multimodal large language models (MLLMs) faces a critical challenge: balancing computational efficiency with the retention of fine-grained spatio-temporal patterns. Existing approaches (e.g., sparse sampling, dense sampling with low resolution, and token compression) suffer from significant information loss in temporal dynamics, spatial details, or subtle interactions, particularly in videos with complex motion or varying resolutions. To address this, we propose Mavors, a novel framework that introduces Multi-granularity video representation for holistic long-video modeling. Specifically, Mavors directly encodes raw video content into latent representations through two core components: 1) an Intra-chunk Vision Encoder (IVE) that preserves high-resolution spatial features via 3D convolutions and Vision Transformers, and 2) an Inter-chunk Feature Aggregator (IFA) that establishes temporal coherence across chunks using transformer-based dependency modeling with chunk-level rotary position encodings. Moreover, the framework unifies image and video understanding by treating images as single-frame videos via sub-image decomposition. Experiments across diverse benchmarks demonstrate Mavors' superiority in maintaining both spatial fidelity and temporal continuity, significantly outperforming existing methods in tasks requiring fine-grained spatio-temporal reasoning.

We present a novel technique for self-supervised video representation learning by: (a) decoupling the learning objective into two contrastive subtasks respectively emphasizing spatial and temporal features, and (b) performing it hierarchically to encourage multi-scale understanding. Motivated by their effectiveness in supervised learning, we first introduce spatial-temporal feature learning decoupling and hierarchical learning to the context of unsupervised video learning. We show by experiments that augmentations can be manipulated as regularization to guide the network to learn desired semantics in contrastive learning, and we propose a way for the model to separately capture spatial and temporal features at multiple scales. We also introduce an approach to overcome the problem of divergent levels of instance invariance at different hierarchies by modeling the invariance as loss weights for objective re-weighting. Experiments on downstream action recognition benchmarks on UCF101 and HMDB51 show that our proposed Hierarchically Decoupled Spatial-Temporal Contrast (HDC) makes substantial improvements over directly learning spatial-temporal features as a whole and achieves competitive performance when compared with other state-of-the-art unsupervised methods. Code will be made available.

Efficient neural representations for dynamic video scenes are critical for applications ranging from video compression to interactive simulations. Yet, existing methods often face challenges related to high memory usage, lengthy training times, and temporal consistency. To address these issues, we introduce a novel neural video representation that combines 3D Gaussian splatting with continuous camera motion modeling. By leveraging Neural ODEs, our approach learns smooth camera trajectories while maintaining an explicit 3D scene representation through Gaussians. Additionally, we introduce a spatiotemporal hierarchical learning strategy, progressively refining spatial and temporal features to enhance reconstruction quality and accelerate convergence. This memory-efficient approach achieves high-quality rendering at impressive speeds. Experimental results show that our hierarchical learning, combined with robust camera motion modeling, captures complex dynamic scenes with strong temporal consistency, achieving state-of-the-art performance across diverse video datasets in both high- and low-motion scenarios.

In a video, an object usually appears as the trajectory, i.e., it spans over a few spatial but longer temporal patches, that contains abundant spatiotemporal contexts. However, modern Video-Language BERTs (VDL-BERTs) neglect this trajectory characteristic that they usually follow image-language BERTs (IL-BERTs) to deploy the patch-to-word (P2W) attention that may over-exploit trivial spatial contexts and neglect significant temporal contexts. To amend this, we propose a novel TW-BERT to learn Trajectory-Word alignment by a newly designed trajectory-to-word (T2W) attention for solving video-language tasks. Moreover, previous VDL-BERTs usually uniformly sample a few frames into the model while different trajectories have diverse graininess, i.e., some trajectories span longer frames and some span shorter, and using a few frames will lose certain useful temporal contexts. However, simply sampling more frames will also make pre-training infeasible due to the largely increased training burdens. To alleviate the problem, during the fine-tuning stage, we insert a novel Hierarchical Frame-Selector (HFS) module into the video encoder. HFS gradually selects the suitable frames conditioned on the text context for the later cross-modal encoder to learn better trajectory-word alignments. By the proposed T2W attention and HFS, our TW-BERT achieves SOTA performances on text-to-video retrieval tasks, and comparable performances on video question-answering tasks with some VDL-BERTs trained on much more data. The code will be available in the supplementary material.

This paper studies the problem of concept-based interpretability of transformer representations for videos. Concretely, we seek to explain the decision-making process of video transformers based on high-level, spatiotemporal concepts that are automatically discovered. Prior research on concept-based interpretability has concentrated solely on image-level tasks. Comparatively, video models deal with the added temporal dimension, increasing complexity and posing challenges in identifying dynamic concepts over time. In this work, we systematically address these challenges by introducing the first Video Transformer Concept Discovery (VTCD) algorithm. To this end, we propose an efficient approach for unsupervised identification of units of video transformer representations - concepts, and ranking their importance to the output of a model. The resulting concepts are highly interpretable, revealing spatio-temporal reasoning mechanisms and object-centric representations in unstructured video models. Performing this analysis jointly over a diverse set of supervised and self-supervised representations, we discover that some of these mechanism are universal in video transformers. Finally, we demonstrate that VTCDcan be used to improve model performance for fine-grained tasks.

Convolutional Neural Networks (CNN) have been regarded as a powerful class of models for image recognition problems. Nevertheless, it is not trivial when utilizing a CNN for learning spatio-temporal video representation. A few studies have shown that performing 3D convolutions is a rewarding approach to capture both spatial and temporal dimensions in videos. However, the development of a very deep 3D CNN from scratch results in expensive computational cost and memory demand. A valid question is why not recycle off-the-shelf 2D networks for a 3D CNN. In this paper, we devise multiple variants of bottleneck building blocks in a residual learning framework by simulating 3times3times3 convolutions with 1times3times3 convolutional filters on spatial domain (equivalent to 2D CNN) plus 3times1times1 convolutions to construct temporal connections on adjacent feature maps in time. Furthermore, we propose a new architecture, named Pseudo-3D Residual Net (P3D ResNet), that exploits all the variants of blocks but composes each in different placement of ResNet, following the philosophy that enhancing structural diversity with going deep could improve the power of neural networks. Our P3D ResNet achieves clear improvements on Sports-1M video classification dataset against 3D CNN and frame-based 2D CNN by 5.3% and 1.8%, respectively. We further examine the generalization performance of video representation produced by our pre-trained P3D ResNet on five different benchmarks and three different tasks, demonstrating superior performances over several state-of-the-art techniques.

Temporal modeling is crucial for various video learning tasks. Most recent approaches employ either factorized (2D+1D) or joint (3D) spatial-temporal operations to extract temporal contexts from the input frames. While the former is more efficient in computation, the latter often obtains better performance. In this paper, we attribute this to a dilemma between the sufficiency and the efficiency of interactions among various positions in different frames. These interactions affect the extraction of task-relevant information shared among frames. To resolve this issue, we prove that frame-by-frame alignments have the potential to increase the mutual information between frame representations, thereby including more task-relevant information to boost effectiveness. Then we propose Alignment-guided Temporal Attention (ATA) to extend 1-dimensional temporal attention with parameter-free patch-level alignments between neighboring frames. It can act as a general plug-in for image backbones to conduct the action recognition task without any model-specific design. Extensive experiments on multiple benchmarks demonstrate the superiority and generality of our module.

The recent advent of Large Language Models (LLMs) has ushered sophisticated reasoning capabilities into the realm of video through Video Large Language Models (VideoLLMs). However, VideoLLMs currently rely on a single vision encoder for all of their visual processing, which limits the amount and type of visual information that can be conveyed to the LLM. Our method, MERV, Multi-Encoder Representation of Videos, instead leverages multiple frozen visual encoders to create a unified representation of a video, providing the VideoLLM with a comprehensive set of specialized visual knowledge. Spatio-temporally aligning the features from each encoder allows us to tackle a wider range of open-ended and multiple-choice video understanding questions and outperform prior state-of-the-art works. MERV is up to 3.7% better in accuracy than Video-LLaVA across the standard suite video understanding benchmarks, while also having a better Video-ChatGPT score. We also improve upon SeViLA, the previous best on zero-shot Perception Test accuracy, by 2.2%. MERV introduces minimal extra parameters and trains faster than equivalent single-encoder methods while parallelizing the visual processing. Finally, we provide qualitative evidence that MERV successfully captures domain knowledge from each of its encoders. Our results offer promising directions in utilizing multiple vision encoders for comprehensive video understanding.

In many robotics and VR/AR applications, 3D-videos are readily-available sources of input (a continuous sequence of depth images, or LIDAR scans). However, those 3D-videos are processed frame-by-frame either through 2D convnets or 3D perception algorithms. In this work, we propose 4-dimensional convolutional neural networks for spatio-temporal perception that can directly process such 3D-videos using high-dimensional convolutions. For this, we adopt sparse tensors and propose the generalized sparse convolution that encompasses all discrete convolutions. To implement the generalized sparse convolution, we create an open-source auto-differentiation library for sparse tensors that provides extensive functions for high-dimensional convolutional neural networks. We create 4D spatio-temporal convolutional neural networks using the library and validate them on various 3D semantic segmentation benchmarks and proposed 4D datasets for 3D-video perception. To overcome challenges in the 4D space, we propose the hybrid kernel, a special case of the generalized sparse convolution, and the trilateral-stationary conditional random field that enforces spatio-temporal consistency in the 7D space-time-chroma space. Experimentally, we show that convolutional neural networks with only generalized 3D sparse convolutions can outperform 2D or 2D-3D hybrid methods by a large margin. Also, we show that on 3D-videos, 4D spatio-temporal convolutional neural networks are robust to noise, outperform 3D convolutional neural networks and are faster than the 3D counterpart in some cases.

We propose a simple, yet effective approach for spatiotemporal feature learning using deep 3-dimensional convolutional networks (3D ConvNets) trained on a large scale supervised video dataset. Our findings are three-fold: 1) 3D ConvNets are more suitable for spatiotemporal feature learning compared to 2D ConvNets; 2) A homogeneous architecture with small 3x3x3 convolution kernels in all layers is among the best performing architectures for 3D ConvNets; and 3) Our learned features, namely C3D (Convolutional 3D), with a simple linear classifier outperform state-of-the-art methods on 4 different benchmarks and are comparable with current best methods on the other 2 benchmarks. In addition, the features are compact: achieving 52.8% accuracy on UCF101 dataset with only 10 dimensions and also very efficient to compute due to the fast inference of ConvNets. Finally, they are conceptually very simple and easy to train and use.

In this paper we discuss several forms of spatiotemporal convolutions for video analysis and study their effects on action recognition. Our motivation stems from the observation that 2D CNNs applied to individual frames of the video have remained solid performers in action recognition. In this work we empirically demonstrate the accuracy advantages of 3D CNNs over 2D CNNs within the framework of residual learning. Furthermore, we show that factorizing the 3D convolutional filters into separate spatial and temporal components yields significantly advantages in accuracy. Our empirical study leads to the design of a new spatiotemporal convolutional block "R(2+1)D" which gives rise to CNNs that achieve results comparable or superior to the state-of-the-art on Sports-1M, Kinetics, UCF101 and HMDB51.

Video understanding requires reasoning at multiple spatiotemporal resolutions -- from short fine-grained motions to events taking place over longer durations. Although transformer architectures have recently advanced the state-of-the-art, they have not explicitly modelled different spatiotemporal resolutions. To this end, we present Multiview Transformers for Video Recognition (MTV). Our model consists of separate encoders to represent different views of the input video with lateral connections to fuse information across views. We present thorough ablation studies of our model and show that MTV consistently performs better than single-view counterparts in terms of accuracy and computational cost across a range of model sizes. Furthermore, we achieve state-of-the-art results on six standard datasets, and improve even further with large-scale pretraining. Code and checkpoints are available at: https://github.com/google-research/scenic/tree/main/scenic/projects/mtv.

Building on the advances of language models, Large Multimodal Models (LMMs) have contributed significant improvements in video understanding. While the current video LMMs utilize advanced Large Language Models (LLMs), they rely on either image or video encoders to process visual inputs, each of which has its own limitations. Image encoders excel at capturing rich spatial details from frame sequences but lack explicit temporal context, which can be important in videos with intricate action sequences. On the other hand, video encoders provide temporal context but are often limited by computational constraints that lead to processing only sparse frames at lower resolutions, resulting in reduced contextual and spatial understanding. To this end, we introduce VideoGPT+, which combines the complementary benefits of the image encoder (for detailed spatial understanding) and the video encoder (for global temporal context modeling). The model processes videos by dividing them into smaller segments and applies an adaptive pooling strategy on features extracted by both image and video encoders. Our architecture showcases improved performance across multiple video benchmarks, including VCGBench, MVBench and Zero-shot question-answering. Further, we develop 112K video-instruction set using a novel semi-automatic annotation pipeline which further improves the model performance. Additionally, to comprehensively evaluate video LMMs, we present VCGBench-Diverse, covering 18 broad video categories such as lifestyle, sports, science, gaming, and surveillance videos. This benchmark with 4,354 question-answer pairs evaluates the generalization of existing LMMs on dense video captioning, spatial and temporal understanding, and complex reasoning, ensuring comprehensive assessment across diverse video types and dynamics. Code: https://github.com/mbzuai-oryx/VideoGPT-plus.

Due to the resource-intensive nature of training vision-language models on expansive video data, a majority of studies have centered on adapting pre-trained image-language models to the video domain. Dominant pipelines propose to tackle the visual discrepancies with additional temporal learners while overlooking the substantial discrepancy for web-scaled descriptive narratives and concise action category names, leading to less distinct semantic space and potential performance limitations. In this work, we prioritize the refinement of text knowledge to facilitate generalizable video recognition. To address the limitations of the less distinct semantic space of category names, we prompt a large language model (LLM) to augment action class names into Spatio-Temporal Descriptors thus bridging the textual discrepancy and serving as a knowledge base for general recognition. Moreover, to assign the best descriptors with different video instances, we propose Optimal Descriptor Solver, forming the video recognition problem as solving the optimal matching flow across frame-level representations and descriptors. Comprehensive evaluations in zero-shot, few-shot, and fully supervised video recognition highlight the effectiveness of our approach. Our best model achieves a state-of-the-art zero-shot accuracy of 75.1% on Kinetics-600.

Video large language models (LLMs) achieve strong video understanding by leveraging a large number of spatio-temporal tokens, but suffer from quadratic computational scaling with token count. To address this, we propose a training-free spatio-temporal token merging method, named STTM. Our key insight is to exploit local spatial and temporal redundancy in video data which has been overlooked in prior work. STTM first transforms each frame into multi-granular spatial tokens using a coarse-to-fine search over a quadtree structure, then performs directed pairwise merging across the temporal dimension. This decomposed merging approach outperforms existing token reduction methods across six video QA benchmarks. Notably, STTM achieves a 2times speed-up with only a 0.5% accuracy drop under a 50% token budget, and a 3times speed-up with just a 2% drop under a 30% budget. Moreover, STTM is query-agnostic, allowing KV cache reuse across different questions for the same video. The project page is available at https://www.jshyun.me/projects/sttm.

Multimodal Large Language Models (MLLMs) have shown promising progress in understanding and analyzing video content. However, processing long videos remains a significant challenge constrained by LLM's context size. To address this limitation, we propose LongVU, a spatiotemporal adaptive compression mechanism thats reduces the number of video tokens while preserving visual details of long videos. Our idea is based on leveraging cross-modal query and inter-frame dependencies to adaptively reduce temporal and spatial redundancy in videos. Specifically, we leverage DINOv2 features to remove redundant frames that exhibit high similarity. Then we utilize text-guided cross-modal query for selective frame feature reduction. Further, we perform spatial token reduction across frames based on their temporal dependencies. Our adaptive compression strategy effectively processes a large number of frames with little visual information loss within given context length. Our LongVU consistently surpass existing methods across a variety of video understanding benchmarks, especially on hour-long video understanding tasks such as VideoMME and MLVU. Given a light-weight LLM, our LongVU also scales effectively into a smaller size with state-of-the-art video understanding performance.

Accurate segmentation of cardiac chambers in echocardiography sequences is crucial for the quantitative analysis of cardiac function, aiding in clinical diagnosis and treatment. The imaging noise, artifacts, and the deformation and motion of the heart pose challenges to segmentation algorithms. While existing methods based on convolutional neural networks, Transformers, and space-time memory networks have improved segmentation accuracy, they often struggle with the trade-off between capturing long-range spatiotemporal dependencies and maintaining computational efficiency with fine-grained feature representation. In this paper, we introduce GDKVM, a novel architecture for echocardiography video segmentation. The model employs Linear Key-Value Association (LKVA) to effectively model inter-frame correlations, and introduces Gated Delta Rule (GDR) to efficiently store intermediate memory states. Key-Pixel Feature Fusion (KPFF) module is designed to integrate local and global features at multiple scales, enhancing robustness against boundary blurring and noise interference. We validated GDKVM on two mainstream echocardiography video datasets (CAMUS and EchoNet-Dynamic) and compared it with various state-of-the-art methods. Experimental results show that GDKVM outperforms existing approaches in terms of segmentation accuracy and robustness, while ensuring real-time performance. Code is available at https://github.com/wangrui2025/GDKVM.

Recently, Spectral Compressive Imaging (SCI) has achieved remarkable success, unlocking significant potential for dynamic spectral vision. However, existing reconstruction methods, primarily image-based, suffer from two limitations: (i) Encoding process masks spatial-spectral features, leading to uncertainty in reconstructing missing information from single compressed measurements, and (ii) The frame-by-frame reconstruction paradigm fails to ensure temporal consistency, which is crucial in the video perception. To address these challenges, this paper seeks to advance spectral reconstruction from the image level to the video level, leveraging the complementary features and temporal continuity across adjacent frames in dynamic scenes. Initially, we construct the first high-quality dynamic hyperspectral image dataset (DynaSpec), comprising 30 sequences obtained through frame-scanning acquisition. Subsequently, we propose the Propagation-Guided Spectral Video Reconstruction Transformer (PG-SVRT), which employs a spatial-then-temporal attention to effectively reconstruct spectral features from abundant video information, while using a bridged token to reduce computational complexity. Finally, we conduct simulation experiments to assess the performance of four SCI systems, and construct a DD-CASSI prototype for real-world data collection and benchmarking. Extensive experiments demonstrate that PG-SVRT achieves superior performance in reconstruction quality, spectral fidelity, and temporal consistency, while maintaining minimal FLOPs. Project page: https://github.com/nju-cite/DynaSpec

Denoising diffusion probabilistic models that were initially proposed for realistic image generation have recently shown success in various perception tasks (e.g., object detection and image segmentation) and are increasingly gaining attention in computer vision. However, extending such models to multi-frame human pose estimation is non-trivial due to the presence of the additional temporal dimension in videos. More importantly, learning representations that focus on keypoint regions is crucial for accurate localization of human joints. Nevertheless, the adaptation of the diffusion-based methods remains unclear on how to achieve such objective. In this paper, we present DiffPose, a novel diffusion architecture that formulates video-based human pose estimation as a conditional heatmap generation problem. First, to better leverage temporal information, we propose SpatioTemporal Representation Learner which aggregates visual evidences across frames and uses the resulting features in each denoising step as a condition. In addition, we present a mechanism called Lookup-based MultiScale Feature Interaction that determines the correlations between local joints and global contexts across multiple scales. This mechanism generates delicate representations that focus on keypoint regions. Altogether, by extending diffusion models, we show two unique characteristics from DiffPose on pose estimation task: (i) the ability to combine multiple sets of pose estimates to improve prediction accuracy, particularly for challenging joints, and (ii) the ability to adjust the number of iterative steps for feature refinement without retraining the model. DiffPose sets new state-of-the-art results on three benchmarks: PoseTrack2017, PoseTrack2018, and PoseTrack21.

AI-generated videos have achieved near-perfect visual realism (e.g., Sora), urgently necessitating reliable detection mechanisms. However, detecting such videos faces significant challenges in modeling high-dimensional spatiotemporal dynamics and identifying subtle anomalies that violate physical laws. In this paper, we propose a physics-driven AI-generated video detection paradigm based on probability flow conservation principles. Specifically, we propose a statistic called Normalized Spatiotemporal Gradient (NSG), which quantifies the ratio of spatial probability gradients to temporal density changes, explicitly capturing deviations from natural video dynamics. Leveraging pre-trained diffusion models, we develop an NSG estimator through spatial gradients approximation and motion-aware temporal modeling without complex motion decomposition while preserving physical constraints. Building on this, we propose an NSG-based video detection method (NSG-VD) that computes the Maximum Mean Discrepancy (MMD) between NSG features of the test and real videos as a detection metric. Last, we derive an upper bound of NSG feature distances between real and generated videos, proving that generated videos exhibit amplified discrepancies due to distributional shifts. Extensive experiments confirm that NSG-VD outperforms state-of-the-art baselines by 16.00% in Recall and 10.75% in F1-Score, validating the superior performance of NSG-VD. The source code is available at https://github.com/ZSHsh98/NSG-VD.

Masked Video Autoencoder (MVA) approaches have demonstrated their potential by significantly outperforming previous video representation learning methods. However, they waste an excessive amount of computations and memory in predicting uninformative tokens/frames due to random masking strategies. (e.g., over 16 nodes with 128 NVIDIA A100 GPUs). To resolve this issue, we exploit the unequal information density among the patches in videos and propose EVEREST, a surprisingly efficient MVA approach for video representation learning that finds tokens containing rich motion features and discards uninformative ones during both pre-training and fine-tuning. We further present an information-intensive frame selection strategy that allows the model to focus on informative and causal frames with minimal redundancy. Our method significantly reduces the computation and memory requirements of MVA, enabling the pre-training and fine-tuning on a single machine with 8 GPUs while achieving comparable performance to computation- and memory-heavy baselines on multiple benchmarks and the uncurated Ego4D dataset. We hope that our work contributes to reducing the barrier to further research on video understanding.

The target of space-time video super-resolution (STVSR) is to increase the spatial-temporal resolution of low-resolution (LR) and low frame rate (LFR) videos. Recent approaches based on deep learning have made significant improvements, but most of them only use two adjacent frames, that is, short-term features, to synthesize the missing frame embedding, which cannot fully explore the information flow of consecutive input LR frames. In addition, existing STVSR models hardly exploit the temporal contexts explicitly to assist high-resolution (HR) frame reconstruction. To address these issues, in this paper, we propose a deformable attention network called STDAN for STVSR. First, we devise a long-short term feature interpolation (LSTFI) module, which is capable of excavating abundant content from more neighboring input frames for the interpolation process through a bidirectional RNN structure. Second, we put forward a spatial-temporal deformable feature aggregation (STDFA) module, in which spatial and temporal contexts in dynamic video frames are adaptively captured and aggregated to enhance SR reconstruction. Experimental results on several datasets demonstrate that our approach outperforms state-of-the-art STVSR methods. The code is available at https://github.com/littlewhitesea/STDAN.

We introduce the task of retrieving relevant video moments from a large corpus of untrimmed, unsegmented videos given a natural language query. Our task poses unique challenges as a system must efficiently identify both the relevant videos and localize the relevant moments in the videos. To address these challenges, we propose SpatioTemporal Alignment with Language (STAL), a model that represents a video moment as a set of regions within a series of short video clips and aligns a natural language query to the moment's regions. Our alignment cost compares variable-length language and video features using symmetric squared Chamfer distance, which allows for efficient indexing and retrieval of the video moments. Moreover, aligning language features to regions within a video moment allows for finer alignment compared to methods that extract only an aggregate feature from the entire video moment. We evaluate our approach on two recently proposed datasets for temporal localization of moments in video with natural language (DiDeMo and Charades-STA) extended to our video corpus moment retrieval setting. We show that our STAL re-ranking model outperforms the recently proposed Moment Context Network on all criteria across all datasets on our proposed task, obtaining relative gains of 37% - 118% for average recall and up to 30% for median rank. Moreover, our approach achieves more than 130x faster retrieval and 8x smaller index size with a 1M video corpus in an approximate setting.

We wish to automatically predict the "speediness" of moving objects in videos---whether they move faster, at, or slower than their "natural" speed. The core component in our approach is SpeedNet---a novel deep network trained to detect if a video is playing at normal rate, or if it is sped up. SpeedNet is trained on a large corpus of natural videos in a self-supervised manner, without requiring any manual annotations. We show how this single, binary classification network can be used to detect arbitrary rates of speediness of objects. We demonstrate prediction results by SpeedNet on a wide range of videos containing complex natural motions, and examine the visual cues it utilizes for making those predictions. Importantly, we show that through predicting the speed of videos, the model learns a powerful and meaningful space-time representation that goes beyond simple motion cues. We demonstrate how those learned features can boost the performance of self-supervised action recognition, and can be used for video retrieval. Furthermore, we also apply SpeedNet for generating time-varying, adaptive video speedups, which can allow viewers to watch videos faster, but with less of the jittery, unnatural motions typical to videos that are sped up uniformly.

By converting low-frame-rate, low-resolution videos into high-frame-rate, high-resolution ones, space-time video super-resolution techniques can enhance visual experiences and facilitate more efficient information dissemination. We propose a convolutional neural network (CNN) for space-time video super-resolution, namely GIRNet. To generate highly accurate features and thus improve performance, the proposed network integrates a feature-level temporal interpolation module with deformable convolutions and a global spatial-temporal information-based residual convolutional long short-term memory (convLSTM) module. In the feature-level temporal interpolation module, we leverage deformable convolution, which adapts to deformations and scale variations of objects across different scene locations. This presents a more efficient solution than conventional convolution for extracting features from moving objects. Our network effectively uses forward and backward feature information to determine inter-frame offsets, leading to the direct generation of interpolated frame features. In the global spatial-temporal information-based residual convLSTM module, the first convLSTM is used to derive global spatial-temporal information from the input features, and the second convLSTM uses the previously computed global spatial-temporal information feature as its initial cell state. This second convLSTM adopts residual connections to preserve spatial information, thereby enhancing the output features. Experiments on the Vimeo90K dataset show that the proposed method outperforms state-of-the-art techniques in peak signal-to-noise-ratio (by 1.45 dB, 1.14 dB, and 0.02 dB over STARnet, TMNet, and 3DAttGAN, respectively), structural similarity index(by 0.027, 0.023, and 0.006 over STARnet, TMNet, and 3DAttGAN, respectively), and visually.

We present a convolution-free approach to video classification built exclusively on self-attention over space and time. Our method, named "TimeSformer," adapts the standard Transformer architecture to video by enabling spatiotemporal feature learning directly from a sequence of frame-level patches. Our experimental study compares different self-attention schemes and suggests that "divided attention," where temporal attention and spatial attention are separately applied within each block, leads to the best video classification accuracy among the design choices considered. Despite the radically new design, TimeSformer achieves state-of-the-art results on several action recognition benchmarks, including the best reported accuracy on Kinetics-400 and Kinetics-600. Finally, compared to 3D convolutional networks, our model is faster to train, it can achieve dramatically higher test efficiency (at a small drop in accuracy), and it can also be applied to much longer video clips (over one minute long). Code and models are available at: https://github.com/facebookresearch/TimeSformer.

The correlation between the vision and text is essential for video moment retrieval (VMR), however, existing methods heavily rely on separate pre-training feature extractors for visual and textual understanding. Without sufficient temporal boundary annotations, it is non-trivial to learn universal video-text alignments. In this work, we explore multi-modal correlations derived from large-scale image-text data to facilitate generalisable VMR. To address the limitations of image-text pre-training models on capturing the video changes, we propose a generic method, referred to as Visual-Dynamic Injection (VDI), to empower the model's understanding of video moments. Whilst existing VMR methods are focusing on building temporal-aware video features, being aware of the text descriptions about the temporal changes is also critical but originally overlooked in pre-training by matching static images with sentences. Therefore, we extract visual context and spatial dynamic information from video frames and explicitly enforce their alignments with the phrases describing video changes (e.g. verb). By doing so, the potentially relevant visual and motion patterns in videos are encoded in the corresponding text embeddings (injected) so to enable more accurate video-text alignments. We conduct extensive experiments on two VMR benchmark datasets (Charades-STA and ActivityNet-Captions) and achieve state-of-the-art performances. Especially, VDI yields notable advantages when being tested on the out-of-distribution splits where the testing samples involve novel scenes and vocabulary.

This paper presents VideoLoom, a unified Video Large Language Model (Video LLM) for joint spatial-temporal understanding. To facilitate the development of fine-grained spatial and temporal localization capabilities, we curate LoomData-8.7k, a human-centric video dataset with temporally grounded and spatially localized captions. With this, VideoLoom achieves state-of-the-art or highly competitive performance across a variety of spatial and temporal benchmarks (e.g., 63.1 J&F on ReVOS for referring video object segmentation, and 48.3 R1@0.7 on Charades-STA for temporal grounding). In addition, we introduce LoomBench, a novel benchmark consisting of temporal, spatial, and compositional video-question pairs, enabling a comprehensive evaluation of Video LLMs from diverse aspects. Collectively, these contributions offer a universal and effective suite for joint spatial-temporal video understanding, setting a new standard in multimodal intelligence.