Hugging Face's logo

qualcomm
/
LiteHRNet

Keypoint Detection
PyTorch
LiteRT
ONNX
android
Model card Files Community
qaihm-bot commited on
Commit
1ea3cba
·
verified ·
1 Parent(s): 4b9cd9c

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +19 -19
README.md CHANGED
@@ -33,34 +33,33 @@ More details on model performance across various devices, can be found
33
 
34
  | Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
35
  |---|---|---|---|---|---|---|---|---|
36
- | LiteHRNet | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 7.847 ms | 0 - 14 MB | FP16 | NPU | [LiteHRNet.tflite](https://huggingface.co/qualcomm/LiteHRNet/blob/main/LiteHRNet.tflite) |
37
- | LiteHRNet | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 7.631 ms | 0 - 23 MB | FP16 | NPU | [LiteHRNet.onnx](https://huggingface.co/qualcomm/LiteHRNet/blob/main/LiteHRNet.onnx) |
38
- | LiteHRNet | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 4.787 ms | 0 - 41 MB | FP16 | NPU | [LiteHRNet.tflite](https://huggingface.co/qualcomm/LiteHRNet/blob/main/LiteHRNet.tflite) |
39
- | LiteHRNet | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 4.719 ms | 0 - 48 MB | FP16 | NPU | [LiteHRNet.onnx](https://huggingface.co/qualcomm/LiteHRNet/blob/main/LiteHRNet.onnx) |
40
- | LiteHRNet | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 5.264 ms | 0 - 37 MB | FP16 | NPU | [LiteHRNet.tflite](https://huggingface.co/qualcomm/LiteHRNet/blob/main/LiteHRNet.tflite) |
41
- | LiteHRNet | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 5.076 ms | 1 - 44 MB | FP16 | NPU | [LiteHRNet.onnx](https://huggingface.co/qualcomm/LiteHRNet/blob/main/LiteHRNet.onnx) |
42
- | LiteHRNet | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 7.88 ms | 0 - 15 MB | FP16 | NPU | [LiteHRNet.tflite](https://huggingface.co/qualcomm/LiteHRNet/blob/main/LiteHRNet.tflite) |
43
- | LiteHRNet | SA7255P ADP | SA7255P | TFLITE | 28.641 ms | 0 - 30 MB | FP16 | NPU | [LiteHRNet.tflite](https://huggingface.co/qualcomm/LiteHRNet/blob/main/LiteHRNet.tflite) |
44
- | LiteHRNet | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 7.839 ms | 0 - 15 MB | FP16 | NPU | [LiteHRNet.tflite](https://huggingface.co/qualcomm/LiteHRNet/blob/main/LiteHRNet.tflite) |
45
- | LiteHRNet | SA8295P ADP | SA8295P | TFLITE | 9.895 ms | 0 - 33 MB | FP16 | NPU | [LiteHRNet.tflite](https://huggingface.co/qualcomm/LiteHRNet/blob/main/LiteHRNet.tflite) |
46
- | LiteHRNet | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 7.86 ms | 0 - 16 MB | FP16 | NPU | [LiteHRNet.tflite](https://huggingface.co/qualcomm/LiteHRNet/blob/main/LiteHRNet.tflite) |
47
- | LiteHRNet | SA8775P ADP | SA8775P | TFLITE | 10.681 ms | 0 - 30 MB | FP16 | NPU | [LiteHRNet.tflite](https://huggingface.co/qualcomm/LiteHRNet/blob/main/LiteHRNet.tflite) |
48
- | LiteHRNet | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 8.46 ms | 0 - 35 MB | FP16 | NPU | [LiteHRNet.tflite](https://huggingface.co/qualcomm/LiteHRNet/blob/main/LiteHRNet.tflite) |
49
- | LiteHRNet | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 8.134 ms | 5 - 5 MB | FP16 | NPU | [LiteHRNet.onnx](https://huggingface.co/qualcomm/LiteHRNet/blob/main/LiteHRNet.onnx) |
50
 
51
 
52
 
53
 
54
  ## Installation
55
 
56
- This model can be installed as a Python package via pip.
57
 
 
58
  ```bash
59
  pip install "qai-hub-models[litehrnet]"
60
  ```
61
 
62
 
63
-
64
  ## Configure Qualcomm® AI Hub to run this model on a cloud-hosted device
65
 
66
  Sign-in to [Qualcomm® AI Hub](https://app.aihub.qualcomm.com/) with your
@@ -112,7 +111,7 @@ LiteHRNet
112
  Device : Samsung Galaxy S23 (13)
113
  Runtime : TFLITE
114
  Estimated inference time (ms) : 7.8
115
- Estimated peak memory usage (MB): [0, 14]
116
  Total # Ops : 1235
117
  Compute Unit(s) : NPU (1233 ops) CPU (2 ops)
118
  ```
@@ -139,7 +138,7 @@ from qai_hub_models.models.litehrnet import Model
139
  torch_model = Model.from_pretrained()
140
 
141
  # Device
142
- device = hub.Device("Samsung Galaxy S23")
143
 
144
  # Trace model
145
  input_shape = torch_model.get_input_spec()
@@ -231,7 +230,8 @@ Explore all available models on [Qualcomm® AI Hub](https://aihub.qualcomm.com/)
231
 
232
 
233
  ## License
234
- * The license for the original implementation of LiteHRNet can be found [here](https://github.com/HRNet/Lite-HRNet/blob/hrnet/LICENSE).
 
235
  * The license for the compiled assets for on-device deployment can be found [here](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/Qualcomm+AI+Hub+Proprietary+License.pdf)
236
 
237
 
 
33
 
34
  | Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
35
  |---|---|---|---|---|---|---|---|---|
36
+ | LiteHRNet | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 7.85 ms | 0 - 16 MB | FP16 | NPU | [LiteHRNet.tflite](https://huggingface.co/qualcomm/LiteHRNet/blob/main/LiteHRNet.tflite) |
37
+ | LiteHRNet | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 7.38 ms | 0 - 24 MB | FP16 | NPU | [LiteHRNet.onnx](https://huggingface.co/qualcomm/LiteHRNet/blob/main/LiteHRNet.onnx) |
38
+ | LiteHRNet | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 4.775 ms | 0 - 36 MB | FP16 | NPU | [LiteHRNet.tflite](https://huggingface.co/qualcomm/LiteHRNet/blob/main/LiteHRNet.tflite) |
39
+ | LiteHRNet | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 4.507 ms | 0 - 48 MB | FP16 | NPU | [LiteHRNet.onnx](https://huggingface.co/qualcomm/LiteHRNet/blob/main/LiteHRNet.onnx) |
40
+ | LiteHRNet | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 5.265 ms | 0 - 37 MB | FP16 | NPU | [LiteHRNet.tflite](https://huggingface.co/qualcomm/LiteHRNet/blob/main/LiteHRNet.tflite) |
41
+ | LiteHRNet | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 4.116 ms | 1 - 45 MB | FP16 | NPU | [LiteHRNet.onnx](https://huggingface.co/qualcomm/LiteHRNet/blob/main/LiteHRNet.onnx) |
42
+ | LiteHRNet | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 7.849 ms | 0 - 15 MB | FP16 | NPU | [LiteHRNet.tflite](https://huggingface.co/qualcomm/LiteHRNet/blob/main/LiteHRNet.tflite) |
43
+ | LiteHRNet | SA7255P ADP | SA7255P | TFLITE | 28.427 ms | 0 - 30 MB | FP16 | NPU | [LiteHRNet.tflite](https://huggingface.co/qualcomm/LiteHRNet/blob/main/LiteHRNet.tflite) |
44
+ | LiteHRNet | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 7.847 ms | 0 - 16 MB | FP16 | NPU | [LiteHRNet.tflite](https://huggingface.co/qualcomm/LiteHRNet/blob/main/LiteHRNet.tflite) |
45
+ | LiteHRNet | SA8295P ADP | SA8295P | TFLITE | 9.954 ms | 0 - 33 MB | FP16 | NPU | [LiteHRNet.tflite](https://huggingface.co/qualcomm/LiteHRNet/blob/main/LiteHRNet.tflite) |
46
+ | LiteHRNet | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 7.871 ms | 0 - 16 MB | FP16 | NPU | [LiteHRNet.tflite](https://huggingface.co/qualcomm/LiteHRNet/blob/main/LiteHRNet.tflite) |
47
+ | LiteHRNet | SA8775P ADP | SA8775P | TFLITE | 10.701 ms | 0 - 30 MB | FP16 | NPU | [LiteHRNet.tflite](https://huggingface.co/qualcomm/LiteHRNet/blob/main/LiteHRNet.tflite) |
48
+ | LiteHRNet | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 8.497 ms | 0 - 31 MB | FP16 | NPU | [LiteHRNet.tflite](https://huggingface.co/qualcomm/LiteHRNet/blob/main/LiteHRNet.tflite) |
49
+ | LiteHRNet | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 8.206 ms | 6 - 6 MB | FP16 | NPU | [LiteHRNet.onnx](https://huggingface.co/qualcomm/LiteHRNet/blob/main/LiteHRNet.onnx) |
50
 
51
 
52
 
53
 
54
  ## Installation
55
 
 
56
 
57
+ Install the package via pip:
58
  ```bash
59
  pip install "qai-hub-models[litehrnet]"
60
  ```
61
 
62
 
 
63
  ## Configure Qualcomm® AI Hub to run this model on a cloud-hosted device
64
 
65
  Sign-in to [Qualcomm® AI Hub](https://app.aihub.qualcomm.com/) with your
 
111
  Device : Samsung Galaxy S23 (13)
112
  Runtime : TFLITE
113
  Estimated inference time (ms) : 7.8
114
+ Estimated peak memory usage (MB): [0, 16]
115
  Total # Ops : 1235
116
  Compute Unit(s) : NPU (1233 ops) CPU (2 ops)
117
  ```
 
138
  torch_model = Model.from_pretrained()
139
 
140
  # Device
141
+ device = hub.Device("Samsung Galaxy S24")
142
 
143
  # Trace model
144
  input_shape = torch_model.get_input_spec()
 
230
 
231
 
232
  ## License
233
+ * The license for the original implementation of LiteHRNet can be found
234
+ [here](https://github.com/HRNet/Lite-HRNet/blob/hrnet/LICENSE).
235
  * The license for the compiled assets for on-device deployment can be found [here](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/Qualcomm+AI+Hub+Proprietary+License.pdf)
236
 
237