Mask Generation
Transformers
Safetensors
sam2
sam2_video
feature-extraction
libreyolo
promptable-segmentation
image-segmentation
Instructions to use LibreYOLO/LibreSAM2tiny with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use LibreYOLO/LibreSAM2tiny with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("mask-generation", model="LibreYOLO/LibreSAM2tiny")# Load model directly from transformers import AutoProcessor, AutoModel processor = AutoProcessor.from_pretrained("LibreYOLO/LibreSAM2tiny") model = AutoModel.from_pretrained("LibreYOLO/LibreSAM2tiny") - sam2
How to use LibreYOLO/LibreSAM2tiny with sam2:
# Use SAM2 with images import torch from sam2.sam2_image_predictor import SAM2ImagePredictor predictor = SAM2ImagePredictor.from_pretrained(LibreYOLO/LibreSAM2tiny) with torch.inference_mode(), torch.autocast("cuda", dtype=torch.bfloat16): predictor.set_image(<your_image>) masks, _, _ = predictor.predict(<input_prompts>)# Use SAM2 with videos import torch from sam2.sam2_video_predictor import SAM2VideoPredictor predictor = SAM2VideoPredictor.from_pretrained(LibreYOLO/LibreSAM2tiny) with torch.inference_mode(), torch.autocast("cuda", dtype=torch.bfloat16): state = predictor.init_state(<your_video>) # add new prompts and instantly get the output on the same frame frame_idx, object_ids, masks = predictor.add_new_points(state, <your_prompts>): # propagate the prompts to get masklets throughout the video for frame_idx, object_ids, masks in predictor.propagate_in_video(state): ... - Notebooks
- Google Colab
- Kaggle
File size: 1,070 Bytes
c30b06e | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 | ---
license: apache-2.0
library_name: transformers
pipeline_tag: mask-generation
tags:
- libreyolo
- sam2
- promptable-segmentation
- image-segmentation
base_model: facebook/sam2.1-hiera-tiny
---
# LibreSAM2tiny
SAM-2.1 Hiera Tiny rehosted for LibreYOLO's `LibreSAM` promptable segmentation tier.
## Source
Derived from [`facebook/sam2.1-hiera-tiny`](https://huggingface.co/facebook/sam2.1-hiera-tiny) at commit
`de431c4043854a71d8101e17995dfe596bf101a5` and the Apache-2.0
[`facebookresearch/sam2`](https://github.com/facebookresearch/sam2) source
release.
## Modifications
Learned parameters are unchanged. The upstream Transformers-compatible snapshot
files are mirrored here for LibreYOLO distribution. This repository adds
LibreYOLO model-card packaging plus `LICENSE` and `NOTICE` files for Apache-2.0
redistribution.
## Usage
```python
from libreyolo import LibreSAM
model = LibreSAM("sam2-tiny")
result = model("image.jpg", points=[500, 375], labels=[1])
```
## License
Apache License 2.0. See [`LICENSE`](./LICENSE) and [`NOTICE`](./NOTICE).
|