Hugging Face's logo

qualcomm
/
Movenet

Keypoint Detection
PyTorch
LiteRT
android
Model card Files Community
qaihm-bot commited on
Commit
d3c7a69
·
verified ·
1 Parent(s): 7d0501b

v0.30.5

Browse files

See https://github.com/quic/ai-hub-models/releases/v0.30.5 for changelog.

Files changed (1) hide show
  1. README.md +24 -24
README.md CHANGED
@@ -34,26 +34,26 @@ More details on model performance across various devices, can be found
34
 
35
  | Model | Precision | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit | Target Model
36
  |---|---|---|---|---|---|---|---|---|
37
- | Movenet | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 7.676 ms | 1 - 10 MB | CPU | [Movenet.tflite](https://huggingface.co/qualcomm/Movenet/blob/main/Movenet.tflite) |
38
- | Movenet | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 8.702 ms | 1 - 24 MB | CPU | [Movenet.tflite](https://huggingface.co/qualcomm/Movenet/blob/main/Movenet.tflite) |
39
- | Movenet | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN | 21.933 ms | 13 - 31 MB | CPU | Use Export Script |
40
- | Movenet | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 3.925 ms | 1 - 4 MB | CPU | [Movenet.tflite](https://huggingface.co/qualcomm/Movenet/blob/main/Movenet.tflite) |
41
- | Movenet | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 5.489 ms | 1 - 15 MB | CPU | [Movenet.tflite](https://huggingface.co/qualcomm/Movenet/blob/main/Movenet.tflite) |
42
- | Movenet | float | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 7.676 ms | 1 - 10 MB | CPU | [Movenet.tflite](https://huggingface.co/qualcomm/Movenet/blob/main/Movenet.tflite) |
43
- | Movenet | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 3.778 ms | 0 - 3 MB | CPU | [Movenet.tflite](https://huggingface.co/qualcomm/Movenet/blob/main/Movenet.tflite) |
44
- | Movenet | float | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 3.835 ms | 1 - 18 MB | CPU | [Movenet.tflite](https://huggingface.co/qualcomm/Movenet/blob/main/Movenet.tflite) |
45
- | Movenet | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 4.423 ms | 1 - 4 MB | CPU | [Movenet.tflite](https://huggingface.co/qualcomm/Movenet/blob/main/Movenet.tflite) |
46
- | Movenet | float | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 5.489 ms | 1 - 15 MB | CPU | [Movenet.tflite](https://huggingface.co/qualcomm/Movenet/blob/main/Movenet.tflite) |
47
- | Movenet | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | TFLITE | 4.174 ms | 1 - 3 MB | CPU | [Movenet.tflite](https://huggingface.co/qualcomm/Movenet/blob/main/Movenet.tflite) |
48
- | Movenet | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | QNN | 20.689 ms | 13 - 15 MB | CPU | Use Export Script |
49
- | Movenet | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | ONNX | 11.21 ms | 7 - 17 MB | CPU | [Movenet.onnx](https://huggingface.co/qualcomm/Movenet/blob/main/Movenet.onnx) |
50
- | Movenet | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 3.362 ms | 0 - 20 MB | CPU | [Movenet.tflite](https://huggingface.co/qualcomm/Movenet/blob/main/Movenet.tflite) |
51
- | Movenet | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN | 11.019 ms | 12 - 31 MB | CPU | Use Export Script |
52
- | Movenet | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 9.412 ms | 6 - 28 MB | CPU | [Movenet.onnx](https://huggingface.co/qualcomm/Movenet/blob/main/Movenet.onnx) |
53
- | Movenet | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | TFLITE | 2.609 ms | 1 - 15 MB | CPU | [Movenet.tflite](https://huggingface.co/qualcomm/Movenet/blob/main/Movenet.tflite) |
54
- | Movenet | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | QNN | 7.469 ms | 12 - 27 MB | CPU | Use Export Script |
55
- | Movenet | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | ONNX | 9.161 ms | 9 - 23 MB | CPU | [Movenet.onnx](https://huggingface.co/qualcomm/Movenet/blob/main/Movenet.onnx) |
56
- | Movenet | float | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 10.135 ms | 18 - 18 MB | CPU | [Movenet.onnx](https://huggingface.co/qualcomm/Movenet/blob/main/Movenet.onnx) |
57
 
58
 
59
 
@@ -117,8 +117,8 @@ Profiling Results
117
  Movenet
118
  Device : cs_8275 (ANDROID 14)
119
  Runtime : TFLITE
120
- Estimated inference time (ms) : 7.7
121
- Estimated peak memory usage (MB): [1, 10]
122
  Total # Ops : 152
123
  Compute Unit(s) : npu (0 ops) gpu (0 ops) cpu (152 ops)
124
  ```
@@ -207,13 +207,13 @@ AI Hub. [Sign up for access](https://myaccount.qualcomm.com/signup).
207
  You can also run the demo on-device.
208
 
209
  ```bash
210
- python -m qai_hub_models.models.movenet.demo --on-device
211
  ```
212
 
213
  **NOTE**: If you want running in a Jupyter Notebook or Google Colab like
214
  environment, please add the following to your cell (instead of the above).
215
  ```
216
- %run -m qai_hub_models.models.movenet.demo -- --on-device
217
  ```
218
 
219
 
 
34
 
35
  | Model | Precision | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit | Target Model
36
  |---|---|---|---|---|---|---|---|---|
37
+ | Movenet | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 7.553 ms | 1 - 11 MB | CPU | [Movenet.tflite](https://huggingface.co/qualcomm/Movenet/blob/main/Movenet.tflite) |
38
+ | Movenet | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 5.658 ms | 1 - 24 MB | CPU | [Movenet.tflite](https://huggingface.co/qualcomm/Movenet/blob/main/Movenet.tflite) |
39
+ | Movenet | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN | 18.54 ms | 13 - 35 MB | CPU | Use Export Script |
40
+ | Movenet | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 3.849 ms | 1 - 4 MB | CPU | [Movenet.tflite](https://huggingface.co/qualcomm/Movenet/blob/main/Movenet.tflite) |
41
+ | Movenet | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 5.492 ms | 1 - 11 MB | CPU | [Movenet.tflite](https://huggingface.co/qualcomm/Movenet/blob/main/Movenet.tflite) |
42
+ | Movenet | float | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 7.553 ms | 1 - 11 MB | CPU | [Movenet.tflite](https://huggingface.co/qualcomm/Movenet/blob/main/Movenet.tflite) |
43
+ | Movenet | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 3.859 ms | 0 - 4 MB | CPU | [Movenet.tflite](https://huggingface.co/qualcomm/Movenet/blob/main/Movenet.tflite) |
44
+ | Movenet | float | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 3.588 ms | 0 - 16 MB | CPU | [Movenet.tflite](https://huggingface.co/qualcomm/Movenet/blob/main/Movenet.tflite) |
45
+ | Movenet | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 3.924 ms | 1 - 4 MB | CPU | [Movenet.tflite](https://huggingface.co/qualcomm/Movenet/blob/main/Movenet.tflite) |
46
+ | Movenet | float | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 5.492 ms | 1 - 11 MB | CPU | [Movenet.tflite](https://huggingface.co/qualcomm/Movenet/blob/main/Movenet.tflite) |
47
+ | Movenet | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | TFLITE | 3.883 ms | 1 - 4 MB | CPU | [Movenet.tflite](https://huggingface.co/qualcomm/Movenet/blob/main/Movenet.tflite) |
48
+ | Movenet | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | QNN | 41.989 ms | 13 - 15 MB | CPU | Use Export Script |
49
+ | Movenet | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | ONNX | 12.176 ms | 0 - 10 MB | CPU | [Movenet.onnx](https://huggingface.co/qualcomm/Movenet/blob/main/Movenet.onnx) |
50
+ | Movenet | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 3.317 ms | 0 - 19 MB | CPU | [Movenet.tflite](https://huggingface.co/qualcomm/Movenet/blob/main/Movenet.tflite) |
51
+ | Movenet | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN | 12.706 ms | 13 - 32 MB | CPU | Use Export Script |
52
+ | Movenet | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 9.361 ms | 8 - 31 MB | CPU | [Movenet.onnx](https://huggingface.co/qualcomm/Movenet/blob/main/Movenet.onnx) |
53
+ | Movenet | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | TFLITE | 2.863 ms | 7 - 21 MB | CPU | [Movenet.tflite](https://huggingface.co/qualcomm/Movenet/blob/main/Movenet.tflite) |
54
+ | Movenet | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | QNN | 7.226 ms | 18 - 32 MB | CPU | Use Export Script |
55
+ | Movenet | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | ONNX | 8.3 ms | 10 - 25 MB | CPU | [Movenet.onnx](https://huggingface.co/qualcomm/Movenet/blob/main/Movenet.onnx) |
56
+ | Movenet | float | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 7.859 ms | 17 - 17 MB | CPU | [Movenet.onnx](https://huggingface.co/qualcomm/Movenet/blob/main/Movenet.onnx) |
57
 
58
 
59
 
 
117
  Movenet
118
  Device : cs_8275 (ANDROID 14)
119
  Runtime : TFLITE
120
+ Estimated inference time (ms) : 7.6
121
+ Estimated peak memory usage (MB): [1, 11]
122
  Total # Ops : 152
123
  Compute Unit(s) : npu (0 ops) gpu (0 ops) cpu (152 ops)
124
  ```
 
207
  You can also run the demo on-device.
208
 
209
  ```bash
210
+ python -m qai_hub_models.models.movenet.demo --eval-mode on-device
211
  ```
212
 
213
  **NOTE**: If you want running in a Jupyter Notebook or Google Colab like
214
  environment, please add the following to your cell (instead of the above).
215
  ```
216
+ %run -m qai_hub_models.models.movenet.demo -- --eval-mode on-device
217
  ```
218
 
219