Keypoint Detection
PyTorch
android
qaihm-bot commited on
Commit
6b95b01
·
verified ·
1 Parent(s): 1b02495

See https://github.com/qualcomm/ai-hub-models/releases/v0.67.0 for changelog.

Files changed (2) hide show
  1. README.md +48 -46
  2. release_assets.json +8 -7
README.md CHANGED
@@ -14,7 +14,7 @@ pipeline_tag: keypoint-detection
14
  LiteHRNet is a machine learning model that detects human pose and returns a location and confidence for each of 17 joints.
15
 
16
  This is based on the implementation of LiteHRNet found [here](https://github.com/HRNet/Lite-HRNet).
17
- This repository contains pre-exported model files optimized for Qualcomm® devices. You can use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/v0.66.0/src/qai_hub_models/models/litehrnet) library to export with custom configurations. More details on model performance across various devices, can be found [here](#performance-summary).
18
 
19
  Qualcomm AI Hub Models uses [Qualcomm AI Hub Workbench](https://workbench.aihub.qualcomm.com) to compile, profile, and evaluate this model. [Sign up](https://myaccount.qualcomm.com/signup) to run these models on a hosted Qualcomm® device.
20
 
@@ -27,23 +27,23 @@ Below are pre-exported model assets ready for deployment.
27
 
28
  | Runtime | Precision | Chipset | SDK Versions | Download |
29
  |---|---|---|---|---|
30
- | ONNX | float | Universal | QAIRT 2.50, ONNX Runtime 1.30.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/litehrnet/releases/v0.66.0/litehrnet-onnx-float.zip)
31
- | QNN_DLC | float | Universal | QAIRT 2.50 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/litehrnet/releases/v0.66.0/litehrnet-qnn_dlc-float.zip)
32
- | TFLITE | float | Universal | QAIRT 2.50 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/litehrnet/releases/v0.66.0/litehrnet-tflite-float.zip)
33
 
34
  For more device-specific assets and performance metrics, visit **[LiteHRNet on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/litehrnet)**.
35
 
36
 
37
  ### Option 2: Export with Custom Configurations
38
 
39
- Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/v0.66.0/src/qai_hub_models/models/litehrnet) Python library to compile and export the model with your own:
40
  - Custom weights (e.g., fine-tuned checkpoints)
41
  - Custom input shapes
42
  - Target device and runtime configurations
43
 
44
  This option is ideal if you need to customize the model beyond the default configuration provided here.
45
 
46
- See our repository for [LiteHRNet on GitHub](https://github.com/qualcomm/ai-hub-models/blob/v0.66.0/src/qai_hub_models/models/litehrnet) for usage instructions.
47
 
48
  ## Model Details
49
 
@@ -57,49 +57,51 @@ See our repository for [LiteHRNet on GitHub](https://github.com/qualcomm/ai-hub-
57
  ## Performance Summary
58
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
59
  |---|---|---|---|---|---|---
60
- | LiteHRNet | ONNX | float | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 2.315 ms | 1 - 102 MB | NPU
61
- | LiteHRNet | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 2.559 ms | 0 - 100 MB | NPU
62
- | LiteHRNet | ONNX | float | Snapdragon® X2 Elite | 2.412 ms | 2 - 2 MB | NPU
63
- | LiteHRNet | ONNX | float | Snapdragon® X Elite | 5.085 ms | 5 - 5 MB | NPU
64
  | LiteHRNet | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 2.986 ms | 0 - 125 MB | NPU
65
- | LiteHRNet | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 5.948 ms | 1 - 122 MB | NPU
66
- | LiteHRNet | ONNX | float | Qualcomm® Dragonwing™ IQ-8275 | 4.347 ms | 1 - 5 MB | NPU
67
- | LiteHRNet | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 4.81 ms | 0 - 8 MB | NPU
68
- | LiteHRNet | ONNX | float | Qualcomm® QCS8450 | 5.948 ms | 1 - 122 MB | NPU
69
- | LiteHRNet | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 5.2 ms | 1 - 4 MB | NPU
70
- | LiteHRNet | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 5.085 ms | 5 - 5 MB | NPU
71
- | LiteHRNet | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 2.559 ms | 0 - 100 MB | NPU
72
- | LiteHRNet | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 0.908 ms | 1 - 85 MB | NPU
73
- | LiteHRNet | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 1.074 ms | 0 - 78 MB | NPU
74
- | LiteHRNet | QNN_DLC | float | Snapdragon® X2 Elite | 1.31 ms | 1 - 1 MB | NPU
75
- | LiteHRNet | QNN_DLC | float | Snapdragon® X Elite | 2.41 ms | 1 - 1 MB | NPU
 
 
76
  | LiteHRNet | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 1.395 ms | 0 - 100 MB | NPU
77
- | LiteHRNet | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 2.911 ms | 0 - 100 MB | NPU
78
- | LiteHRNet | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 2.181 ms | 1 - 4 MB | NPU
79
- | LiteHRNet | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 2.091 ms | 1 - 2 MB | NPU
80
- | LiteHRNet | QNN_DLC | float | Qualcomm® SA8775P | 2.68 ms | 1 - 79 MB | NPU
81
- | LiteHRNet | QNN_DLC | float | Qualcomm® SA8650P | 2.68 ms | 1 - 79 MB | NPU
82
- | LiteHRNet | QNN_DLC | float | Qualcomm® SA8255P | 2.68 ms | 1 - 79 MB | NPU
83
- | LiteHRNet | QNN_DLC | float | Qualcomm® QCS8450 | 2.911 ms | 0 - 100 MB | NPU
84
- | LiteHRNet | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 2.476 ms | 1 - 3 MB | NPU
85
- | LiteHRNet | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 2.41 ms | 1 - 1 MB | NPU
86
- | LiteHRNet | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 1.074 ms | 0 - 78 MB | NPU
87
- | LiteHRNet | QNN_DLC | float | Qualcomm® SA7255P | 4.981 ms | 1 - 78 MB | NPU
88
- | LiteHRNet | QNN_DLC | float | Qualcomm® SA8295P | 3.348 ms | 0 - 76 MB | NPU
89
- | LiteHRNet | TFLITE | float | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 2.049 ms | 0 - 109 MB | NPU
90
- | LiteHRNet | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 2.225 ms | 0 - 106 MB | NPU
91
  | LiteHRNet | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 2.673 ms | 0 - 146 MB | NPU
92
- | LiteHRNet | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 5.224 ms | 0 - 135 MB | NPU
93
- | LiteHRNet | TFLITE | float | Qualcomm® Dragonwing™ IQ-8275 | 4.198 ms | 1 - 12 MB | NPU
94
- | LiteHRNet | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 4.255 ms | 1 - 5 MB | NPU
95
- | LiteHRNet | TFLITE | float | Qualcomm® SA8775P | 5.247 ms | 1 - 112 MB | NPU
96
- | LiteHRNet | TFLITE | float | Qualcomm® SA8650P | 5.247 ms | 1 - 112 MB | NPU
97
- | LiteHRNet | TFLITE | float | Qualcomm® SA8255P | 5.247 ms | 1 - 112 MB | NPU
98
- | LiteHRNet | TFLITE | float | Qualcomm® QCS8450 | 5.224 ms | 0 - 135 MB | NPU
99
- | LiteHRNet | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 4.91 ms | 1 - 12 MB | NPU
100
- | LiteHRNet | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 2.225 ms | 0 - 106 MB | NPU
101
- | LiteHRNet | TFLITE | float | Qualcomm® SA7255P | 8.564 ms | 1 - 110 MB | NPU
102
- | LiteHRNet | TFLITE | float | Qualcomm® SA8295P | 6.011 ms | 1 - 111 MB | NPU
 
 
103
 
104
  ## License
105
  * The license for the original implementation of LiteHRNet can be found
 
14
  LiteHRNet is a machine learning model that detects human pose and returns a location and confidence for each of 17 joints.
15
 
16
  This is based on the implementation of LiteHRNet found [here](https://github.com/HRNet/Lite-HRNet).
17
+ This repository contains pre-exported model files optimized for Qualcomm® devices. You can use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/v0.67.0/src/qai_hub_models/models/litehrnet) library to export with custom configurations. More details on model performance across various devices, can be found [here](#performance-summary).
18
 
19
  Qualcomm AI Hub Models uses [Qualcomm AI Hub Workbench](https://workbench.aihub.qualcomm.com) to compile, profile, and evaluate this model. [Sign up](https://myaccount.qualcomm.com/signup) to run these models on a hosted Qualcomm® device.
20
 
 
27
 
28
  | Runtime | Precision | Chipset | SDK Versions | Download |
29
  |---|---|---|---|---|
30
+ | ONNX | float | Universal | QAIRT 2.51, ONNX Runtime 1.30.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/litehrnet/releases/v0.67.0/litehrnet-onnx-float.zip)
31
+ | QNN_DLC | float | Universal | QAIRT 2.51 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/litehrnet/releases/v0.67.0/litehrnet-qnn_dlc-float.zip)
32
+ | TFLITE | float | Universal | QAIRT 2.51 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/litehrnet/releases/v0.67.0/litehrnet-tflite-float.zip)
33
 
34
  For more device-specific assets and performance metrics, visit **[LiteHRNet on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/litehrnet)**.
35
 
36
 
37
  ### Option 2: Export with Custom Configurations
38
 
39
+ Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/v0.67.0/src/qai_hub_models/models/litehrnet) Python library to compile and export the model with your own:
40
  - Custom weights (e.g., fine-tuned checkpoints)
41
  - Custom input shapes
42
  - Target device and runtime configurations
43
 
44
  This option is ideal if you need to customize the model beyond the default configuration provided here.
45
 
46
+ See our repository for [LiteHRNet on GitHub](https://github.com/qualcomm/ai-hub-models/blob/v0.67.0/src/qai_hub_models/models/litehrnet) for usage instructions.
47
 
48
  ## Model Details
49
 
 
57
  ## Performance Summary
58
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
59
  |---|---|---|---|---|---|---
60
+ | LiteHRNet | ONNX | float | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 1.968 ms | 0 - 106 MB | NPU
61
+ | LiteHRNet | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 2.215 ms | 0 - 105 MB | NPU
62
+ | LiteHRNet | ONNX | float | Snapdragon® X2 Elite | 2.069 ms | 2 - 2 MB | NPU
63
+ | LiteHRNet | ONNX | float | Snapdragon® X Elite | 4.499 ms | 6 - 6 MB | NPU
64
  | LiteHRNet | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 2.986 ms | 0 - 125 MB | NPU
65
+ | LiteHRNet | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 5.164 ms | 1 - 127 MB | NPU
66
+ | LiteHRNet | ONNX | float | Qualcomm® Dragonwing™ IQ-8275 | 3.688 ms | 1 - 5 MB | NPU
67
+ | LiteHRNet | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 4.229 ms | 0 - 8 MB | NPU
68
+ | LiteHRNet | ONNX | float | Qualcomm® QCS8450 | 5.164 ms | 1 - 127 MB | NPU
69
+ | LiteHRNet | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 4.404 ms | 1 - 4 MB | NPU
70
+ | LiteHRNet | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 4.499 ms | 6 - 6 MB | NPU
71
+ | LiteHRNet | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 2.215 ms | 0 - 105 MB | NPU
72
+ | LiteHRNet | ONNX | float | Snapdragon® 8 Elite Extreme Gen 6 Mobile | 1.778 ms | 1 - 103 MB | NPU
73
+ | LiteHRNet | ONNX | float | Snapdragon® 8 Elite Gen 6 Mobile | 2.095 ms | 0 - 101 MB | NPU
74
+ | LiteHRNet | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 1.826 ms | 1 - 90 MB | NPU
75
+ | LiteHRNet | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 2.118 ms | 1 - 91 MB | NPU
76
+ | LiteHRNet | QNN_DLC | float | Snapdragon® X2 Elite | 2.218 ms | 1 - 1 MB | NPU
77
+ | LiteHRNet | QNN_DLC | float | Snapdragon® X Elite | 4.515 ms | 1 - 1 MB | NPU
78
  | LiteHRNet | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 1.395 ms | 0 - 100 MB | NPU
79
+ | LiteHRNet | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 5.061 ms | 0 - 108 MB | NPU
80
+ | LiteHRNet | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 3.493 ms | 1 - 4 MB | NPU
81
+ | LiteHRNet | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 4.067 ms | 1 - 2 MB | NPU
82
+ | LiteHRNet | QNN_DLC | float | Qualcomm® QCS8450 | 5.061 ms | 0 - 108 MB | NPU
83
+ | LiteHRNet | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 4.37 ms | 1 - 4 MB | NPU
84
+ | LiteHRNet | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 4.515 ms | 1 - 1 MB | NPU
85
+ | LiteHRNet | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 2.118 ms | 1 - 91 MB | NPU
86
+ | LiteHRNet | QNN_DLC | float | Qualcomm® SA8295P | 5.409 ms | 1 - 85 MB | NPU
87
+ | LiteHRNet | QNN_DLC | float | Snapdragon® 8 Elite Extreme Gen 6 Mobile | 1.716 ms | 0 - 85 MB | NPU
88
+ | LiteHRNet | QNN_DLC | float | Snapdragon® 8 Elite Gen 6 Mobile | 2.01 ms | 0 - 85 MB | NPU
89
+ | LiteHRNet | TFLITE | float | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 1.536 ms | 0 - 81 MB | NPU
90
+ | LiteHRNet | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 1.709 ms | 0 - 85 MB | NPU
 
 
91
  | LiteHRNet | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 2.673 ms | 0 - 146 MB | NPU
92
+ | LiteHRNet | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 3.957 ms | 1 - 103 MB | NPU
93
+ | LiteHRNet | TFLITE | float | Qualcomm® Dragonwing™ IQ-8275 | 3.128 ms | 1 - 12 MB | NPU
94
+ | LiteHRNet | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 3.353 ms | 0 - 2 MB | NPU
95
+ | LiteHRNet | TFLITE | float | Qualcomm® SA8775P | 4.346 ms | 1 - 81 MB | NPU
96
+ | LiteHRNet | TFLITE | float | Qualcomm® SA8650P | 4.346 ms | 1 - 81 MB | NPU
97
+ | LiteHRNet | TFLITE | float | Qualcomm® SA8255P | 4.346 ms | 1 - 81 MB | NPU
98
+ | LiteHRNet | TFLITE | float | Qualcomm® QCS8450 | 3.957 ms | 1 - 103 MB | NPU
99
+ | LiteHRNet | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 3.731 ms | 1 - 12 MB | NPU
100
+ | LiteHRNet | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 1.709 ms | 0 - 85 MB | NPU
101
+ | LiteHRNet | TFLITE | float | Qualcomm® SA7255P | 20.985 ms | 4 - 17 MB | CPU
102
+ | LiteHRNet | TFLITE | float | Qualcomm® SA8295P | 4.787 ms | 1 - 82 MB | NPU
103
+ | LiteHRNet | TFLITE | float | Snapdragon® 8 Elite Extreme Gen 6 Mobile | 1.427 ms | 0 - 83 MB | NPU
104
+ | LiteHRNet | TFLITE | float | Snapdragon® 8 Elite Gen 6 Mobile | 1.653 ms | 0 - 84 MB | NPU
105
 
106
  ## License
107
  * The license for the original implementation of LiteHRNet can be found
release_assets.json CHANGED
@@ -1,26 +1,27 @@
1
  {
2
- "version": "0.66.0",
3
  "precisions": {
4
  "float": {
5
  "universal_assets": {
6
  "onnx": {
7
  "tool_versions": {
8
- "qairt": "2.50.0.260828221209",
9
  "onnx_runtime": "1.30.0"
10
  },
11
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/litehrnet/releases/v0.66.0/litehrnet-onnx-float.zip"
12
  },
13
  "qnn_dlc": {
14
  "tool_versions": {
15
- "qairt": "2.50.0.260828221209"
16
  },
17
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/litehrnet/releases/v0.66.0/litehrnet-qnn_dlc-float.zip"
18
  },
19
  "tflite": {
20
  "tool_versions": {
21
- "qairt": "2.50.0.260828221209"
 
22
  },
23
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/litehrnet/releases/v0.66.0/litehrnet-tflite-float.zip"
24
  }
25
  }
26
  }
 
1
  {
2
+ "version": "0.67.0",
3
  "precisions": {
4
  "float": {
5
  "universal_assets": {
6
  "onnx": {
7
  "tool_versions": {
8
+ "qairt": "2.51.0.260929181845",
9
  "onnx_runtime": "1.30.0"
10
  },
11
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/litehrnet/releases/v0.67.0/litehrnet-onnx-float.zip"
12
  },
13
  "qnn_dlc": {
14
  "tool_versions": {
15
+ "qairt": "2.51.0.260929181845"
16
  },
17
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/litehrnet/releases/v0.67.0/litehrnet-qnn_dlc-float.zip"
18
  },
19
  "tflite": {
20
  "tool_versions": {
21
+ "qairt": "2.51.0.260929181845",
22
+ "litert": "1.4.2"
23
  },
24
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/litehrnet/releases/v0.67.0/litehrnet-tflite-float.zip"
25
  }
26
  }
27
  }