Hugging Face's logo

qualcomm
/
HRNet-W48-OCR

Image Segmentation
PyTorch
bu_auto
real_time
android
Model card Files
xet
 Community
qaihm-bot commited on
Commit
6895c2f
·
verified ·
1 Parent(s): 3bde88f

v0.63.0

Browse files

See https://github.com/qualcomm/ai-hub-models/releases/v0.63.0 for changelog.

Files changed (2) hide show
  1. README.md +32 -16
  2. release_assets.json +14 -3
README.md CHANGED
@@ -16,7 +16,7 @@ pipeline_tag: image-segmentation
16
  HRNet-W48-OCR is a machine learning model that can segment images from the Cityscape dataset. It has lightweight and hardware-efficient operations and thus delivers significant speedup on diverse hardware platforms
17
 
18
  This is based on the implementation of HRNet-W48-OCR found [here](https://github.com/HRNet/HRNet-Semantic-Segmentation).
19
- 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/v0.62.2/src/qai_hub_models/models/hrnet_w48_ocr) library to export with custom configurations. More details on model performance across various devices, can be found [here](#performance-summary).
20
 
21
  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.
22
 
@@ -29,21 +29,22 @@ Below are pre-exported model assets ready for deployment.
29
 
30
  | Runtime | Precision | Chipset | SDK Versions | Download |
31
  |---|---|---|---|---|
32
- | ONNX | float | Universal | QAIRT 2.45, ONNX Runtime 1.27.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/hrnet_w48_ocr/releases/v0.62.2/hrnet_w48_ocr-onnx-float.zip)
 
33
 
34
  For more device-specific assets and performance metrics, visit **[HRNet-W48-OCR on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/hrnet_w48_ocr)**.
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/v0.62.2/src/qai_hub_models/models/hrnet_w48_ocr) 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 [HRNet-W48-OCR on GitHub](https://github.com/qualcomm/ai-hub-models/blob/v0.62.2/src/qai_hub_models/models/hrnet_w48_ocr) for usage instructions.
47
 
48
  ## Model Details
49
 
@@ -60,18 +61,33 @@ See our repository for [HRNet-W48-OCR on GitHub](https://github.com/qualcomm/ai-
60
  ## Performance Summary
61
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
62
  |---|---|---|---|---|---|---
63
- | HRNet-W48-OCR | ONNX | float | Snapdragon® X2 Elite | 621.034 ms | 33 - 33 MB | NPU
64
- | HRNet-W48-OCR | ONNX | float | Snapdragon® X Elite | 994.037 ms | 134 - 134 MB | NPU
65
- | HRNet-W48-OCR | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 785.319 ms | 25 - 2295 MB | NPU
66
- | HRNet-W48-OCR | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 2062.471 ms | 28 - 2174 MB | NPU
67
- | HRNet-W48-OCR | ONNX | float | Qualcomm® Dragonwing™ IQ-8275 | 1229.607 ms | 24 - 51 MB | NPU
68
- | HRNet-W48-OCR | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1094.759 ms | 0 - 136 MB | NPU
69
- | HRNet-W48-OCR | ONNX | float | Qualcomm® QCS8450 | 2062.471 ms | 28 - 2174 MB | NPU
70
- | HRNet-W48-OCR | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 1579.811 ms | 24 - 51 MB | NPU
71
- | HRNet-W48-OCR | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 994.037 ms | 134 - 134 MB | NPU
72
- | HRNet-W48-OCR | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 692.815 ms | 12 - 1131 MB | NPU
73
- | HRNet-W48-OCR | ONNX | float | Snapdragon® 8 Elite Mobile | 692.815 ms | 12 - 1131 MB | NPU
74
- | HRNet-W48-OCR | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 541.494 ms | 12 - 1236 MB | NPU
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
75
 
76
  ## License
77
  * The license for the original implementation of HRNet-W48-OCR can be found
 
16
  HRNet-W48-OCR is a machine learning model that can segment images from the Cityscape dataset. It has lightweight and hardware-efficient operations and thus delivers significant speedup on diverse hardware platforms
17
 
18
  This is based on the implementation of HRNet-W48-OCR found [here](https://github.com/HRNet/HRNet-Semantic-Segmentation).
19
+ 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/v0.63.0/src/qai_hub_models/models/hrnet_w48_ocr) library to export with custom configurations. More details on model performance across various devices, can be found [here](#performance-summary).
20
 
21
  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.
22
 
 
29
 
30
  | Runtime | Precision | Chipset | SDK Versions | Download |
31
  |---|---|---|---|---|
32
+ | ONNX | float | Universal | QAIRT 2.50, ONNX Runtime 1.27.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/hrnet_w48_ocr/releases/v0.63.0/hrnet_w48_ocr-onnx-float.zip)
33
+ | ONNX | w8a16 | Universal | QAIRT 2.50, ONNX Runtime 1.27.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/hrnet_w48_ocr/releases/v0.63.0/hrnet_w48_ocr-onnx-w8a16.zip)
34
 
35
  For more device-specific assets and performance metrics, visit **[HRNet-W48-OCR on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/hrnet_w48_ocr)**.
36
 
37
 
38
  ### Option 2: Export with Custom Configurations
39
 
40
+ Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/v0.63.0/src/qai_hub_models/models/hrnet_w48_ocr) Python library to compile and export the model with your own:
41
  - Custom weights (e.g., fine-tuned checkpoints)
42
  - Custom input shapes
43
  - Target device and runtime configurations
44
 
45
  This option is ideal if you need to customize the model beyond the default configuration provided here.
46
 
47
+ See our repository for [HRNet-W48-OCR on GitHub](https://github.com/qualcomm/ai-hub-models/blob/v0.63.0/src/qai_hub_models/models/hrnet_w48_ocr) for usage instructions.
48
 
49
  ## Model Details
50
 
 
61
  ## Performance Summary
62
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
63
  |---|---|---|---|---|---|---
64
+ | HRNet-W48-OCR | ONNX | float | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 462.282 ms | 21 - 1139 MB | NPU
65
+ | HRNet-W48-OCR | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 639.752 ms | 12 - 1063 MB | NPU
66
+ | HRNet-W48-OCR | ONNX | float | Snapdragon® X2 Elite | 540.091 ms | 33 - 33 MB | NPU
67
+ | HRNet-W48-OCR | ONNX | float | Snapdragon® X Elite | 998.821 ms | 132 - 132 MB | NPU
68
+ | HRNet-W48-OCR | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 678.269 ms | 25 - 2219 MB | NPU
69
+ | HRNet-W48-OCR | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 1456.968 ms | 26 - 2133 MB | NPU
70
+ | HRNet-W48-OCR | ONNX | float | Qualcomm® Dragonwing™ IQ-8275 | 1142.534 ms | 24 - 52 MB | NPU
71
+ | HRNet-W48-OCR | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 935.261 ms | 24 - 28 MB | NPU
72
+ | HRNet-W48-OCR | ONNX | float | Qualcomm® QCS8450 | 1456.968 ms | 26 - 2133 MB | NPU
73
+ | HRNet-W48-OCR | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 1135.83 ms | 24 - 51 MB | NPU
74
+ | HRNet-W48-OCR | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 998.821 ms | 132 - 132 MB | NPU
75
+ | HRNet-W48-OCR | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 639.752 ms | 12 - 1063 MB | NPU
76
+ | HRNet-W48-OCR | ONNX | w8a16 | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 413.187 ms | 13 - 2014 MB | NPU
77
+ | HRNet-W48-OCR | ONNX | w8a16 | Snapdragon® 8 Elite For Galaxy Mobile | 532.471 ms | 6 - 1801 MB | NPU
78
+ | HRNet-W48-OCR | ONNX | w8a16 | Snapdragon® X2 Elite | 452.128 ms | 17 - 17 MB | NPU
79
+ | HRNet-W48-OCR | ONNX | w8a16 | Snapdragon® X Elite | 748.732 ms | 79 - 79 MB | NPU
80
+ | HRNet-W48-OCR | ONNX | w8a16 | Snapdragon® 8 Gen 3 Mobile | 631.084 ms | 0 - 2608 MB | NPU
81
+ | HRNet-W48-OCR | ONNX | w8a16 | Snapdragon® 8 Gen 1 Mobile | 1018.392 ms | 7 - 2663 MB | NPU
82
+ | HRNet-W48-OCR | ONNX | w8a16 | Qualcomm® Dragonwing™ QCS6490 | 2944.892 ms | 0 - 16 MB | NPU
83
+ | HRNet-W48-OCR | ONNX | w8a16 | Qualcomm® Dragonwing™ IQ-8275 | 740.31 ms | 5 - 21 MB | NPU
84
+ | HRNet-W48-OCR | ONNX | w8a16 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 832.832 ms | 0 - 86 MB | NPU
85
+ | HRNet-W48-OCR | ONNX | w8a16 | Qualcomm® QCS8450 | 1018.392 ms | 7 - 2663 MB | NPU
86
+ | HRNet-W48-OCR | ONNX | w8a16 | Qualcomm® Dragonwing™ IQ-9075 | 898.195 ms | 9 - 24 MB | NPU
87
+ | HRNet-W48-OCR | ONNX | w8a16 | Qualcomm® Dragonwing™ IQ-X7181 | 748.732 ms | 79 - 79 MB | NPU
88
+ | HRNet-W48-OCR | ONNX | w8a16 | Qualcomm® Dragonwing™ Q-7790 | 1073.522 ms | 11 - 2296 MB | NPU
89
+ | HRNet-W48-OCR | ONNX | w8a16 | Qualcomm® Dragonwing™ Q-8750 | 532.471 ms | 6 - 1801 MB | NPU
90
+ | HRNet-W48-OCR | ONNX | w8a16 | Snapdragon® 7 Gen 4 Mobile | 1073.522 ms | 11 - 2296 MB | NPU
91
 
92
  ## License
93
  * The license for the original implementation of HRNet-W48-OCR can be found
release_assets.json CHANGED
@@ -1,14 +1,25 @@
1
  {
2
- "version": "0.62.2",
3
  "precisions": {
4
  "float": {
5
  "universal_assets": {
6
  "onnx": {
7
  "tool_versions": {
8
- "qairt": "2.45.0.260326154327",
9
  "onnx_runtime": "1.27.1"
10
  },
11
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/hrnet_w48_ocr/releases/v0.62.2/hrnet_w48_ocr-onnx-float.zip"
 
 
 
 
 
 
 
 
 
 
 
12
  }
13
  }
14
  }
 
1
  {
2
+ "version": "0.63.0",
3
  "precisions": {
4
  "float": {
5
  "universal_assets": {
6
  "onnx": {
7
  "tool_versions": {
8
+ "qairt": "2.50.0.260828221209",
9
  "onnx_runtime": "1.27.1"
10
  },
11
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/hrnet_w48_ocr/releases/v0.63.0/hrnet_w48_ocr-onnx-float.zip"
12
+ }
13
+ }
14
+ },
15
+ "w8a16": {
16
+ "universal_assets": {
17
+ "onnx": {
18
+ "tool_versions": {
19
+ "qairt": "2.50.0.260828221209",
20
+ "onnx_runtime": "1.27.1"
21
+ },
22
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/hrnet_w48_ocr/releases/v0.63.0/hrnet_w48_ocr-onnx-w8a16.zip"
23
  }
24
  }
25
  }