OWL-V2: Optimized for Qualcomm Devices
OWL-V2 (Open-World Localization with Vision Transformers) is an open-vocabulary object detector that uses a CLIP-based ViT-B/16 backbone. Given an image and one or more free-form text queries, the model predicts bounding boxes and confidence scores for each query.
This is based on the implementation of OWL-V2 found here. This repository contains pre-exported model files optimized for Qualcomm® devices. You can use the Qualcomm® AI Hub Models library to export with custom configurations. More details on model performance across various devices, can be found here.
Qualcomm AI Hub Models uses Qualcomm AI Hub Workbench to compile, profile, and evaluate this model. Sign up to run these models on a hosted Qualcomm® device.
Getting Started
There are two ways to deploy this model on your device:
Option 1: Download Pre-Exported Models
Below are pre-exported model assets ready for deployment.
| Runtime | Precision | Chipset | SDK Versions | Download |
|---|---|---|---|---|
| PRECOMPILED_QNN_ONNX | float | Snapdragon® X2 Elite | QAIRT 2.45, ONNX Runtime 1.27.1 | Download |
| PRECOMPILED_QNN_ONNX | float | Snapdragon® X Elite | QAIRT 2.45, ONNX Runtime 1.27.1 | Download |
| PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Gen 3 Mobile | QAIRT 2.45, ONNX Runtime 1.27.1 | Download |
| PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Gen 1 Mobile | QAIRT 2.45, ONNX Runtime 1.27.1 | Download |
| PRECOMPILED_QNN_ONNX | float | Qualcomm® Dragonwing™ IQ-8275 | QAIRT 2.45, ONNX Runtime 1.27.1 | Download |
| PRECOMPILED_QNN_ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | QAIRT 2.45, ONNX Runtime 1.27.1 | Download |
| PRECOMPILED_QNN_ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | QAIRT 2.45, ONNX Runtime 1.27.1 | Download |
| PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite Mobile | QAIRT 2.45, ONNX Runtime 1.27.1 | Download |
| PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | QAIRT 2.45, ONNX Runtime 1.27.1 | Download |
| QNN_CONTEXT_BINARY | float | Snapdragon® X2 Elite | QAIRT 2.45 | Download |
| QNN_CONTEXT_BINARY | float | Snapdragon® X Elite | QAIRT 2.45 | Download |
| QNN_CONTEXT_BINARY | float | Snapdragon® 8 Gen 3 Mobile | QAIRT 2.45 | Download |
| QNN_CONTEXT_BINARY | float | Snapdragon® 8 Gen 1 Mobile | QAIRT 2.45 | Download |
| QNN_CONTEXT_BINARY | float | Qualcomm® Dragonwing™ IQ-8275 | QAIRT 2.45 | Download |
| QNN_CONTEXT_BINARY | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | QAIRT 2.45 | Download |
| QNN_CONTEXT_BINARY | float | Qualcomm® SA8775P | QAIRT 2.45 | Download |
| QNN_CONTEXT_BINARY | float | Qualcomm® Dragonwing™ IQ-9075 | QAIRT 2.45 | Download |
| QNN_CONTEXT_BINARY | float | Qualcomm® SA7255P | QAIRT 2.45 | Download |
| QNN_CONTEXT_BINARY | float | Qualcomm® SA8295P | QAIRT 2.45 | Download |
| QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite Mobile | QAIRT 2.45 | Download |
| QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite Gen 5 Mobile | QAIRT 2.45 | Download |
For more device-specific assets and performance metrics, visit OWL-V2 on Qualcomm® AI Hub.
Option 2: Export with Custom Configurations
Use the Qualcomm® AI Hub Models Python library to compile and export the model with your own:
- Custom weights (e.g., fine-tuned checkpoints)
- Custom input shapes
- Target device and runtime configurations
This option is ideal if you need to customize the model beyond the default configuration provided here.
See our repository for OWL-V2 on GitHub for usage instructions.
Model Details
Model Type: Model_use_case.object_detection
Model Stats:
- Input image resolution: 960x960
- Model checkpoint: google/owlv2-base-patch16-ensemble
- Model size (float): 620 MB
- Number of parameters: 155M
Performance Summary
| Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit |
|---|---|---|---|---|---|---|
| text | PRECOMPILED_QNN_ONNX | float | Snapdragon® X2 Elite | 3.319 ms | 11 - 11 MB | NPU |
| text | PRECOMPILED_QNN_ONNX | float | Snapdragon® X Elite | 6.009 ms | 116 - 116 MB | NPU |
| text | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Gen 3 Mobile | 4.219 ms | 11 - 22 MB | NPU |
| text | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Gen 1 Mobile | 13.246 ms | 11 - 25 MB | NPU |
| text | PRECOMPILED_QNN_ONNX | float | Qualcomm® Dragonwing™ IQ-8275 | 8.01 ms | 10 - 24 MB | NPU |
| text | PRECOMPILED_QNN_ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 5.923 ms | 11 - 13 MB | NPU |
| text | PRECOMPILED_QNN_ONNX | float | Qualcomm® QCS8450 | 13.246 ms | 11 - 25 MB | NPU |
| text | PRECOMPILED_QNN_ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 7.135 ms | 10 - 24 MB | NPU |
| text | PRECOMPILED_QNN_ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 6.009 ms | 116 - 116 MB | NPU |
| text | PRECOMPILED_QNN_ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 3.688 ms | 2 - 16 MB | NPU |
| text | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite Mobile | 3.688 ms | 2 - 16 MB | NPU |
| text | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 3.4 ms | 1 - 8 MB | NPU |
| text | QNN_CONTEXT_BINARY | float | Snapdragon® X2 Elite | 3.76 ms | 11 - 11 MB | NPU |
| text | QNN_CONTEXT_BINARY | float | Snapdragon® X Elite | 6.042 ms | 11 - 11 MB | NPU |
| text | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Gen 3 Mobile | 4.13 ms | 11 - 18 MB | NPU |
| text | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Gen 1 Mobile | 13.224 ms | 11 - 21 MB | NPU |
| text | QNN_CONTEXT_BINARY | float | Qualcomm® Dragonwing™ IQ-8275 | 7.95 ms | 11 - 23 MB | NPU |
| text | QNN_CONTEXT_BINARY | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 5.921 ms | 11 - 12 MB | NPU |
| text | QNN_CONTEXT_BINARY | float | Qualcomm® SA8775P | 7.288 ms | 1 - 10 MB | NPU |
| text | QNN_CONTEXT_BINARY | float | Qualcomm® SA8650P | 7.288 ms | 1 - 10 MB | NPU |
| text | QNN_CONTEXT_BINARY | float | Qualcomm® SA8255P | 7.288 ms | 1 - 10 MB | NPU |
| text | QNN_CONTEXT_BINARY | float | Qualcomm® QCS8450 | 13.224 ms | 11 - 21 MB | NPU |
| text | QNN_CONTEXT_BINARY | float | Qualcomm® Dragonwing™ IQ-9075 | 7.063 ms | 11 - 23 MB | NPU |
| text | QNN_CONTEXT_BINARY | float | Qualcomm® Dragonwing™ IQ-X7181 | 6.042 ms | 11 - 11 MB | NPU |
| text | QNN_CONTEXT_BINARY | float | Qualcomm® Dragonwing™ Q-8750 | 3.643 ms | 0 - 9 MB | NPU |
| text | QNN_CONTEXT_BINARY | float | Qualcomm® SA7255P | 18.904 ms | 1 - 11 MB | NPU |
| text | QNN_CONTEXT_BINARY | float | Qualcomm® SA8295P | 12.428 ms | 0 - 5 MB | NPU |
| text | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite Mobile | 3.643 ms | 0 - 9 MB | NPU |
| text | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite Gen 5 Mobile | 3.404 ms | 11 - 19 MB | NPU |
| vision | PRECOMPILED_QNN_ONNX | float | Snapdragon® X2 Elite | 597.239 ms | 33 - 33 MB | NPU |
| vision | PRECOMPILED_QNN_ONNX | float | Snapdragon® X Elite | 1248.828 ms | 297 - 297 MB | NPU |
| vision | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Gen 3 Mobile | 1030.615 ms | 11 - 23 MB | NPU |
| vision | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Gen 1 Mobile | 2116.625 ms | 12 - 26 MB | NPU |
| vision | PRECOMPILED_QNN_ONNX | float | Qualcomm® Dragonwing™ IQ-8275 | 1229.586 ms | 11 - 25 MB | NPU |
| vision | PRECOMPILED_QNN_ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1396.941 ms | 0 - 310 MB | NPU |
| vision | PRECOMPILED_QNN_ONNX | float | Qualcomm® QCS8450 | 2116.625 ms | 12 - 26 MB | NPU |
| vision | PRECOMPILED_QNN_ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 1208.266 ms | 10 - 24 MB | NPU |
| vision | PRECOMPILED_QNN_ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 1248.828 ms | 297 - 297 MB | NPU |
| vision | PRECOMPILED_QNN_ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 712.532 ms | 24 - 36 MB | NPU |
| vision | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite Mobile | 712.532 ms | 24 - 36 MB | NPU |
| vision | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 655.556 ms | 13 - 25 MB | NPU |
| vision | QNN_CONTEXT_BINARY | float | Snapdragon® X2 Elite | 597.043 ms | 11 - 11 MB | NPU |
| vision | QNN_CONTEXT_BINARY | float | Snapdragon® X Elite | 1222.138 ms | 11 - 11 MB | NPU |
| vision | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Gen 3 Mobile | 965.972 ms | 11 - 19 MB | NPU |
| vision | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Gen 1 Mobile | 2054.581 ms | 11 - 25 MB | NPU |
| vision | QNN_CONTEXT_BINARY | float | Qualcomm® Dragonwing™ IQ-8275 | 1230.145 ms | 11 - 34 MB | NPU |
| vision | QNN_CONTEXT_BINARY | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1406.702 ms | 11 - 12 MB | NPU |
| vision | QNN_CONTEXT_BINARY | float | Qualcomm® SA8775P | 1268.382 ms | 3 - 11 MB | NPU |
| vision | QNN_CONTEXT_BINARY | float | Qualcomm® SA8650P | 1268.382 ms | 3 - 11 MB | NPU |
| vision | QNN_CONTEXT_BINARY | float | Qualcomm® SA8255P | 1268.382 ms | 3 - 11 MB | NPU |
| vision | QNN_CONTEXT_BINARY | float | Qualcomm® QCS8450 | 2054.581 ms | 11 - 25 MB | NPU |
| vision | QNN_CONTEXT_BINARY | float | Qualcomm® Dragonwing™ IQ-9075 | 1208.852 ms | 13 - 35 MB | NPU |
| vision | QNN_CONTEXT_BINARY | float | Qualcomm® Dragonwing™ IQ-X7181 | 1222.138 ms | 11 - 11 MB | NPU |
| vision | QNN_CONTEXT_BINARY | float | Qualcomm® Dragonwing™ Q-8750 | 700.173 ms | 12 - 21 MB | NPU |
| vision | QNN_CONTEXT_BINARY | float | Qualcomm® SA7255P | 2448.088 ms | 1 - 10 MB | NPU |
| vision | QNN_CONTEXT_BINARY | float | Qualcomm® SA8295P | 1429.837 ms | 0 - 5 MB | NPU |
| vision | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite Mobile | 700.173 ms | 12 - 21 MB | NPU |
| vision | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite Gen 5 Mobile | 626.664 ms | 11 - 19 MB | NPU |
License
- The license for the original implementation of OWL-V2 can be found here.
References
Community
- Join our AI Hub Slack community to collaborate, post questions and learn more about on-device AI.
- For questions or feedback please reach out to us.
