Hugging Face's logo

qualcomm
/
Track-Anything

PyTorch
android
video-object-tracking
Model card Files
xet
 Community

YAML Metadata Warning:The pipeline tag "video-object-tracking" is not in the official list: text-classification, token-classification, table-question-answering, question-answering, zero-shot-classification, translation, summarization, feature-extraction, text-generation, fill-mask, sentence-similarity, text-to-speech, text-to-audio, automatic-speech-recognition, audio-to-audio, audio-classification, audio-text-to-text, voice-activity-detection, depth-estimation, image-classification, object-detection, image-segmentation, text-to-image, image-to-text, image-to-image, image-to-video, unconditional-image-generation, video-classification, reinforcement-learning, robotics, tabular-classification, tabular-regression, tabular-to-text, table-to-text, multiple-choice, text-ranking, text-retrieval, time-series-forecasting, text-to-video, image-text-to-text, image-text-to-image, image-text-to-video, visual-question-answering, document-question-answering, zero-shot-image-classification, graph-ml, mask-generation, zero-shot-object-detection, text-to-3d, image-to-3d, image-feature-extraction, video-text-to-text, keypoint-detection, visual-document-retrieval, any-to-any, video-to-video, other

Track-Anything: Optimized for Qualcomm Devices

Track-Anything is a video based machine learning model to track an object in a video.

This is based on the implementation of Track-Anything 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
ONNX float Universal QAIRT 2.45, ONNX Runtime 1.25.0 Download
TFLITE float Universal QAIRT 2.45 Download

For more device-specific assets and performance metrics, visit Track-Anything 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 Track-Anything on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.video_object_tracking

Model Stats:

  • Model checkpoint: xmem
  • Input resolution: 320x568
  • Number of parameters (encode_key_with_shrinkage): 9.72M
  • Model size (encode_key_with_shrinkage) (float): 37.1 MB
  • Number of parameters (encode_value): 23.3M
  • Model size (encode_value) (float): 88.8 MB
  • Number of parameters (encode_key_without_shrinkage): 9.71M
  • Model size (encode_key_without_shrinkage) (float): 37.1 MB
  • Number of parameters (segment): 30.1M
  • Model size (segment) (float): 115 MB

Performance Summary

Model Runtime Precision Chipset Inference Time (ms) Peak Memory Range (MB) Primary Compute Unit
encode_key_with_shrinkage ONNX float Snapdragon® X2 Elite 2.772 ms 211 - 211 MB NPU
encode_key_with_shrinkage ONNX float Snapdragon® X Elite 5.418 ms 179 - 179 MB NPU
encode_key_with_shrinkage ONNX float Snapdragon® 8 Gen 3 Mobile 3.714 ms 1 - 190 MB NPU
encode_key_with_shrinkage ONNX float Snapdragon® 8 Gen 1 Mobile 10.662 ms 9 - 182 MB NPU
encode_key_with_shrinkage ONNX float Qualcomm® QCS8550 (Proxy) 5.198 ms 6 - 40 MB NPU
encode_key_with_shrinkage ONNX float Qualcomm® QCS8450 10.662 ms 9 - 182 MB NPU
encode_key_with_shrinkage ONNX float Snapdragon® 8 Elite Gen 5 Mobile 2.368 ms 0 - 167 MB NPU
encode_key_with_shrinkage ONNX float Qualcomm® QCS9075 9.65 ms 10 - 55 MB NPU
encode_key_with_shrinkage ONNX float Snapdragon® 8 Elite Mobile 3.002 ms 0 - 169 MB NPU
encode_key_with_shrinkage ONNX float Qualcomm® QCS8750 3.002 ms 0 - 169 MB NPU
encode_key_with_shrinkage ONNX float Qualcomm® QCS7181 5.418 ms 179 - 179 MB NPU
encode_key_with_shrinkage TFLITE float Snapdragon® 8 Gen 3 Mobile 4.063 ms 0 - 214 MB NPU
encode_key_with_shrinkage TFLITE float Snapdragon® 8 Gen 1 Mobile 11.757 ms 3 - 200 MB NPU
encode_key_with_shrinkage TFLITE float Qualcomm® QCS8275 31.243 ms 3 - 180 MB NPU
encode_key_with_shrinkage TFLITE float Qualcomm® QCS8550 (Proxy) 5.953 ms 3 - 6 MB NPU
encode_key_with_shrinkage TFLITE float Qualcomm® SA8775P 49.415 ms 3 - 25 MB GPU
encode_key_with_shrinkage TFLITE float Qualcomm® SA8650P 49.415 ms 3 - 25 MB GPU
encode_key_with_shrinkage TFLITE float Qualcomm® SA8255P 49.415 ms 3 - 25 MB GPU
encode_key_with_shrinkage TFLITE float Qualcomm® QCS8450 11.757 ms 3 - 200 MB NPU
encode_key_with_shrinkage TFLITE float Snapdragon® 8 Elite Gen 5 Mobile 2.445 ms 0 - 185 MB NPU
encode_key_with_shrinkage TFLITE float Qualcomm® SA7255P 31.243 ms 3 - 180 MB NPU
encode_key_with_shrinkage TFLITE float Qualcomm® QCS9075 10.671 ms 2 - 29 MB NPU
encode_key_with_shrinkage TFLITE float Snapdragon® 8 Elite Mobile 3.28 ms 0 - 180 MB NPU
encode_key_with_shrinkage TFLITE float Qualcomm® SA8295P 10.297 ms 3 - 163 MB NPU
encode_key_with_shrinkage TFLITE float Qualcomm® QCS8750 3.28 ms 0 - 180 MB NPU
encode_key_without_shrinkage ONNX float Snapdragon® X2 Elite 3.853 ms 211 - 211 MB NPU
encode_key_without_shrinkage ONNX float Snapdragon® X Elite 6.678 ms 179 - 179 MB NPU
encode_key_without_shrinkage ONNX float Snapdragon® 8 Gen 3 Mobile 5.032 ms 29 - 220 MB NPU
encode_key_without_shrinkage ONNX float Snapdragon® 8 Gen 1 Mobile 13.496 ms 26 - 217 MB NPU
encode_key_without_shrinkage ONNX float Qualcomm® QCS8550 (Proxy) 6.893 ms 42 - 63 MB NPU
encode_key_without_shrinkage ONNX float Qualcomm® QCS8450 13.496 ms 26 - 217 MB NPU
encode_key_without_shrinkage ONNX float Snapdragon® 8 Elite Gen 5 Mobile 3.454 ms 0 - 174 MB NPU
encode_key_without_shrinkage ONNX float Qualcomm® QCS9075 11.531 ms 52 - 98 MB NPU
encode_key_without_shrinkage ONNX float Snapdragon® 8 Elite Mobile 4.356 ms 51 - 248 MB NPU
encode_key_without_shrinkage ONNX float Qualcomm® QCS8750 4.356 ms 51 - 248 MB NPU
encode_key_without_shrinkage ONNX float Qualcomm® QCS7181 6.678 ms 179 - 179 MB NPU
encode_key_without_shrinkage TFLITE float Snapdragon® 8 Gen 3 Mobile 5.563 ms 0 - 215 MB NPU
encode_key_without_shrinkage TFLITE float Snapdragon® 8 Gen 1 Mobile 14.243 ms 20 - 221 MB NPU
encode_key_without_shrinkage TFLITE float Qualcomm® QCS8275 33.56 ms 20 - 204 MB NPU
encode_key_without_shrinkage TFLITE float Qualcomm® QCS8550 (Proxy) 7.898 ms 20 - 22 MB NPU
encode_key_without_shrinkage TFLITE float Qualcomm® SA8775P 52.355 ms 20 - 42 MB GPU
encode_key_without_shrinkage TFLITE float Qualcomm® SA8650P 52.355 ms 20 - 42 MB GPU
encode_key_without_shrinkage TFLITE float Qualcomm® SA8255P 52.355 ms 20 - 42 MB GPU
encode_key_without_shrinkage TFLITE float Qualcomm® QCS8450 14.243 ms 20 - 221 MB NPU
encode_key_without_shrinkage TFLITE float Snapdragon® 8 Elite Gen 5 Mobile 3.558 ms 0 - 221 MB NPU
encode_key_without_shrinkage TFLITE float Qualcomm® SA7255P 33.56 ms 20 - 204 MB NPU
encode_key_without_shrinkage TFLITE float Qualcomm® QCS9075 12.887 ms 20 - 63 MB NPU
encode_key_without_shrinkage TFLITE float Snapdragon® 8 Elite Mobile 4.387 ms 17 - 203 MB NPU
encode_key_without_shrinkage TFLITE float Qualcomm® SA8295P 12.663 ms 20 - 184 MB NPU
encode_key_without_shrinkage TFLITE float Qualcomm® QCS8750 4.387 ms 17 - 203 MB NPU
encode_value ONNX float Snapdragon® X2 Elite 6.541 ms 207 - 207 MB NPU
encode_value ONNX float Snapdragon® X Elite 13.255 ms 143 - 143 MB NPU
encode_value ONNX float Snapdragon® 8 Gen 3 Mobile 11.37 ms 3 - 202 MB NPU
encode_value ONNX float Snapdragon® 8 Gen 1 Mobile 31.64 ms 12 - 196 MB NPU
encode_value ONNX float Qualcomm® QCS8550 (Proxy) 16.747 ms 9 - 13 MB NPU
encode_value ONNX float Qualcomm® QCS8450 31.64 ms 12 - 196 MB NPU
encode_value ONNX float Snapdragon® 8 Elite Gen 5 Mobile 8.225 ms 7 - 176 MB NPU
encode_value ONNX float Qualcomm® QCS9075 18.64 ms 6 - 51 MB NPU
encode_value ONNX float Snapdragon® 8 Elite Mobile 9.544 ms 7 - 192 MB NPU
encode_value ONNX float Qualcomm® QCS8750 9.544 ms 7 - 192 MB NPU
encode_value ONNX float Qualcomm® QCS7181 13.255 ms 143 - 143 MB NPU
encode_value TFLITE float Snapdragon® 8 Gen 3 Mobile 8.226 ms 1 - 287 MB NPU
encode_value TFLITE float Snapdragon® 8 Gen 1 Mobile 21.175 ms 3 - 291 MB NPU
encode_value TFLITE float Qualcomm® QCS8275 51.302 ms 3 - 217 MB NPU
encode_value TFLITE float Qualcomm® QCS8550 (Proxy) 11.89 ms 0 - 3 MB NPU
encode_value TFLITE float Qualcomm® SA8775P 254.547 ms 12 - 30 MB CPU
encode_value TFLITE float Qualcomm® SA8650P 254.547 ms 12 - 30 MB CPU
encode_value TFLITE float Qualcomm® SA8255P 254.547 ms 12 - 30 MB CPU
encode_value TFLITE float Qualcomm® QCS8450 21.175 ms 3 - 291 MB NPU
encode_value TFLITE float Snapdragon® 8 Elite Gen 5 Mobile 5.357 ms 2 - 202 MB NPU
encode_value TFLITE float Qualcomm® SA7255P 51.302 ms 3 - 217 MB NPU
encode_value TFLITE float Qualcomm® QCS9075 19.689 ms 0 - 55 MB NPU
encode_value TFLITE float Snapdragon® 8 Elite Mobile 6.822 ms 3 - 217 MB NPU
encode_value TFLITE float Qualcomm® SA8295P 18.897 ms 3 - 219 MB NPU
encode_value TFLITE float Qualcomm® QCS8750 6.822 ms 3 - 217 MB NPU
segment ONNX float Snapdragon® X2 Elite 11.665 ms 191 - 191 MB NPU
segment ONNX float Snapdragon® X Elite 23.303 ms 161 - 161 MB NPU
segment ONNX float Snapdragon® 8 Gen 3 Mobile 17.472 ms 24 - 259 MB NPU
segment ONNX float Snapdragon® 8 Gen 1 Mobile 42.657 ms 25 - 244 MB NPU
segment ONNX float Qualcomm® QCS8550 (Proxy) 22.914 ms 3 - 59 MB NPU
segment ONNX float Qualcomm® QCS8450 42.657 ms 25 - 244 MB NPU
segment ONNX float Snapdragon® 8 Elite Gen 5 Mobile 10.66 ms 5 - 214 MB NPU
segment ONNX float Qualcomm® QCS9075 38.795 ms 21 - 78 MB NPU
segment ONNX float Snapdragon® 8 Elite Mobile 13.308 ms 3 - 208 MB NPU
segment ONNX float Qualcomm® QCS8750 13.308 ms 3 - 208 MB NPU
segment ONNX float Qualcomm® QCS7181 23.303 ms 161 - 161 MB NPU
segment TFLITE float Snapdragon® 8 Gen 3 Mobile 17.325 ms 2 - 294 MB NPU
segment TFLITE float Snapdragon® 8 Gen 1 Mobile 42.262 ms 2 - 293 MB NPU
segment TFLITE float Qualcomm® QCS8275 103.303 ms 2 - 195 MB NPU
segment TFLITE float Qualcomm® QCS8550 (Proxy) 23.322 ms 2 - 4 MB NPU
segment TFLITE float Qualcomm® SA8775P 647.324 ms 28 - 47 MB CPU
segment TFLITE float Qualcomm® SA8650P 647.324 ms 28 - 47 MB CPU
segment TFLITE float Qualcomm® SA8255P 647.324 ms 28 - 47 MB CPU
segment TFLITE float Qualcomm® QCS8450 42.262 ms 2 - 293 MB NPU
segment TFLITE float Snapdragon® 8 Elite Gen 5 Mobile 11.452 ms 2 - 198 MB NPU
segment TFLITE float Qualcomm® SA7255P 103.303 ms 2 - 195 MB NPU
segment TFLITE float Qualcomm® QCS9075 38.101 ms 0 - 83 MB NPU
segment TFLITE float Snapdragon® 8 Elite Mobile 13.22 ms 1 - 195 MB NPU
segment TFLITE float Qualcomm® SA8295P 37.579 ms 2 - 202 MB NPU
segment TFLITE float Qualcomm® QCS8750 13.22 ms 1 - 195 MB NPU

License

  • The license for the original implementation of Track-Anything can be found here.

References

Community

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Paper for qualcomm/Track-Anything