Hugging Face's logo

qualcomm
/
MaskRCNN

Image Segmentation
PyTorch
android
Model card Files
xet
 Community
qaihm-bot commited on
Commit
06cb7be
·
verified ·
1 Parent(s): e0120b9

v0.62.0

Browse files

See https://github.com/qualcomm/ai-hub-models/releases/v0.62.0 for changelog.

Files changed (2) hide show
  1. README.md +57 -59
  2. release_assets.json +3 -3
README.md CHANGED
@@ -14,7 +14,7 @@ pipeline_tag: image-segmentation
14
  Mask R-CNN is a machine learning model that extends Faster R-CNN to perform instance segmentation by detecting objects in an image while simultaneously generating a high-quality segmentation mask for each instance. It adds a branch for predicting segmentation masks in parallel with the existing branch for bounding box recognition.
15
 
16
  This is based on the implementation of MaskRCNN found [here](https://github.com/pytorch/vision).
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/v0.61.0/src/qai_hub_models/models/maskrcnn) 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
 
@@ -27,22 +27,22 @@ Below are pre-exported model assets ready for deployment.
27
 
28
  | Runtime | Precision | Chipset | SDK Versions | Download |
29
  |---|---|---|---|---|
30
- | 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/maskrcnn/releases/v0.61.0/maskrcnn-onnx-float.zip)
31
- | QNN_DLC | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/maskrcnn/releases/v0.61.0/maskrcnn-qnn_dlc-float.zip)
32
 
33
  For more device-specific assets and performance metrics, visit **[MaskRCNN on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/maskrcnn)**.
34
 
35
 
36
  ### Option 2: Export with Custom Configurations
37
 
38
- Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/v0.61.0/src/qai_hub_models/models/maskrcnn) Python library to compile and export the model with your own:
39
  - Custom weights (e.g., fine-tuned checkpoints)
40
  - Custom input shapes
41
  - Target device and runtime configurations
42
 
43
  This option is ideal if you need to customize the model beyond the default configuration provided here.
44
 
45
- See our repository for [MaskRCNN on GitHub](https://github.com/qualcomm/ai-hub-models/blob/v0.61.0/src/qai_hub_models/models/maskrcnn) for usage instructions.
46
 
47
  ## Model Details
48
 
@@ -58,60 +58,58 @@ See our repository for [MaskRCNN on GitHub](https://github.com/qualcomm/ai-hub-m
58
  ## Performance Summary
59
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
60
  |---|---|---|---|---|---|---
61
- | proposal_generator | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 68.909 ms | 141 - 2954 MB | NPU
62
- | proposal_generator | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 144.135 ms | 130 - 2864 MB | NPU
63
- | proposal_generator | ONNX | float | Qualcomm® Dragonwing™ IQ-8275 | 125.557 ms | 7 - 18 MB | NPU
64
- | proposal_generator | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 92.71 ms | 116 - 310 MB | NPU
65
- | proposal_generator | ONNX | float | Qualcomm® QCS8450 | 144.135 ms | 130 - 2864 MB | NPU
66
- | proposal_generator | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 118.572 ms | 7 - 17 MB | NPU
67
- | proposal_generator | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 53.021 ms | 134 - 1632 MB | NPU
68
- | proposal_generator | ONNX | float | Snapdragon® 8 Elite Mobile | 53.021 ms | 134 - 1632 MB | NPU
69
- | proposal_generator | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 41.455 ms | 134 - 1487 MB | NPU
70
- | proposal_generator | QNN_DLC | float | Snapdragon® X2 Elite | 43.268 ms | 7 - 7 MB | NPU
71
- | proposal_generator | QNN_DLC | float | Snapdragon® X Elite | 91.22 ms | 7 - 7 MB | NPU
72
- | proposal_generator | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 69.747 ms | 88 - 2408 MB | NPU
73
- | proposal_generator | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 151.28 ms | 8 - 2738 MB | NPU
74
- | proposal_generator | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 127.366 ms | 7 - 72 MB | NPU
75
- | proposal_generator | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 353.808 ms | 3 - 1759 MB | NPU
76
- | proposal_generator | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 95.943 ms | 8 - 833 MB | NPU
77
- | proposal_generator | QNN_DLC | float | Qualcomm® SA8775P | 122.288 ms | 2 - 1758 MB | NPU
78
- | proposal_generator | QNN_DLC | float | Qualcomm® SA8650P | 122.288 ms | 2 - 1758 MB | NPU
79
- | proposal_generator | QNN_DLC | float | Qualcomm® SA8255P | 122.288 ms | 2 - 1758 MB | NPU
80
- | proposal_generator | QNN_DLC | float | Qualcomm® QCS8450 | 151.28 ms | 8 - 2738 MB | NPU
81
- | proposal_generator | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 122.991 ms | 7 - 71 MB | NPU
82
- | proposal_generator | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 91.22 ms | 7 - 7 MB | NPU
83
- | proposal_generator | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 53.386 ms | 7 - 1517 MB | NPU
84
- | proposal_generator | QNN_DLC | float | Qualcomm® SA7255P | 353.808 ms | 3 - 1759 MB | NPU
85
- | proposal_generator | QNN_DLC | float | Qualcomm® SA8295P | 126.051 ms | 0 - 1429 MB | NPU
86
- | proposal_generator | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 53.386 ms | 7 - 1517 MB | NPU
87
- | proposal_generator | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 42.056 ms | 7 - 1789 MB | NPU
88
- | roi_head | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 71.751 ms | 239 - 714 MB | NPU
89
- | roi_head | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 140.16 ms | 294 - 740 MB | NPU
90
- | roi_head | ONNX | float | Qualcomm® Dragonwing™ IQ-8275 | 154.982 ms | 108 - 208 MB | NPU
91
- | roi_head | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 90.542 ms | 272 - 275 MB | NPU
92
- | roi_head | ONNX | float | Qualcomm® QCS8450 | 140.16 ms | 294 - 740 MB | NPU
93
- | roi_head | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 137.846 ms | 111 - 210 MB | NPU
94
- | roi_head | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 62.574 ms | 200 - 598 MB | NPU
95
- | roi_head | ONNX | float | Snapdragon® 8 Elite Mobile | 62.574 ms | 200 - 598 MB | NPU
96
- | roi_head | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 50.007 ms | 228 - 626 MB | NPU
97
- | roi_head | QNN_DLC | float | Snapdragon® X2 Elite | 90.198 ms | 52 - 52 MB | NPU
98
- | roi_head | QNN_DLC | float | Snapdragon® X Elite | 225.299 ms | 52 - 52 MB | NPU
99
- | roi_head | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 162.031 ms | 13 - 800 MB | NPU
100
- | roi_head | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 293.869 ms | 39 - 789 MB | NPU
101
- | roi_head | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 242.731 ms | 52 - 107 MB | NPU
102
- | roi_head | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 553.268 ms | 41 - 546 MB | NPU
103
- | roi_head | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 243.856 ms | 50 - 52 MB | NPU
104
- | roi_head | QNN_DLC | float | Qualcomm® SA8775P | 256.613 ms | 46 - 775 MB | NPU
105
- | roi_head | QNN_DLC | float | Qualcomm® SA8650P | 256.613 ms | 46 - 775 MB | NPU
106
- | roi_head | QNN_DLC | float | Qualcomm® SA8255P | 256.613 ms | 46 - 775 MB | NPU
107
- | roi_head | QNN_DLC | float | Qualcomm® QCS8450 | 293.869 ms | 39 - 789 MB | NPU
108
- | roi_head | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 255.554 ms | 54 - 108 MB | NPU
109
- | roi_head | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 225.299 ms | 52 - 52 MB | NPU
110
- | roi_head | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 112.298 ms | 21 - 511 MB | NPU
111
- | roi_head | QNN_DLC | float | Qualcomm® SA7255P | 553.268 ms | 41 - 546 MB | NPU
112
- | roi_head | QNN_DLC | float | Qualcomm® SA8295P | 288.077 ms | 49 - 770 MB | NPU
113
- | roi_head | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 112.298 ms | 21 - 511 MB | NPU
114
- | roi_head | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 82.119 ms | 15 - 515 MB | NPU
115
 
116
  ## License
117
  * The license for the original implementation of MaskRCNN can be found
 
14
  Mask R-CNN is a machine learning model that extends Faster R-CNN to perform instance segmentation by detecting objects in an image while simultaneously generating a high-quality segmentation mask for each instance. It adds a branch for predicting segmentation masks in parallel with the existing branch for bounding box recognition.
15
 
16
  This is based on the implementation of MaskRCNN found [here](https://github.com/pytorch/vision).
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/v0.62.0/src/qai_hub_models/models/maskrcnn) 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 2.45, ONNX Runtime 1.27.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/maskrcnn/releases/v0.62.0/maskrcnn-onnx-float.zip)
31
+ | QNN_DLC | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/maskrcnn/releases/v0.62.0/maskrcnn-qnn_dlc-float.zip)
32
 
33
  For more device-specific assets and performance metrics, visit **[MaskRCNN on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/maskrcnn)**.
34
 
35
 
36
  ### Option 2: Export with Custom Configurations
37
 
38
+ Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/v0.62.0/src/qai_hub_models/models/maskrcnn) Python library to compile and export the model with your own:
39
  - Custom weights (e.g., fine-tuned checkpoints)
40
  - Custom input shapes
41
  - Target device and runtime configurations
42
 
43
  This option is ideal if you need to customize the model beyond the default configuration provided here.
44
 
45
+ See our repository for [MaskRCNN on GitHub](https://github.com/qualcomm/ai-hub-models/blob/v0.62.0/src/qai_hub_models/models/maskrcnn) for usage instructions.
46
 
47
  ## Model Details
48
 
 
58
  ## Performance Summary
59
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
60
  |---|---|---|---|---|---|---
61
+ | proposal_generator | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 70.231 ms | 0 - 2812 MB | NPU
62
+ | proposal_generator | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 142.534 ms | 130 - 2863 MB | NPU
63
+ | proposal_generator | ONNX | float | Qualcomm® Dragonwing™ IQ-8275 | 124.965 ms | 24 - 35 MB | NPU
64
+ | proposal_generator | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 92.531 ms | 123 - 199 MB | NPU
65
+ | proposal_generator | ONNX | float | Qualcomm® QCS8450 | 142.534 ms | 130 - 2863 MB | NPU
66
+ | proposal_generator | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 119.396 ms | 7 - 17 MB | NPU
67
+ | proposal_generator | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 52.601 ms | 127 - 1623 MB | NPU
68
+ | proposal_generator | ONNX | float | Snapdragon® 8 Elite Mobile | 52.601 ms | 127 - 1623 MB | NPU
69
+ | proposal_generator | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 41.473 ms | 126 - 1479 MB | NPU
70
+ | proposal_generator | QNN_DLC | float | Snapdragon® X2 Elite | 42.762 ms | 7 - 7 MB | NPU
71
+ | proposal_generator | QNN_DLC | float | Snapdragon® X Elite | 92.038 ms | 7 - 7 MB | NPU
72
+ | proposal_generator | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 69.972 ms | 7 - 2330 MB | NPU
73
+ | proposal_generator | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 147.799 ms | 8 - 2739 MB | NPU
74
+ | proposal_generator | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 127.626 ms | 7 - 72 MB | NPU
75
+ | proposal_generator | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 94.57 ms | 7 - 10 MB | NPU
76
+ | proposal_generator | QNN_DLC | float | Qualcomm® SA8775P | 122.238 ms | 1 - 1758 MB | NPU
77
+ | proposal_generator | QNN_DLC | float | Qualcomm® SA8650P | 122.238 ms | 1 - 1758 MB | NPU
78
+ | proposal_generator | QNN_DLC | float | Qualcomm® SA8255P | 122.238 ms | 1 - 1758 MB | NPU
79
+ | proposal_generator | QNN_DLC | float | Qualcomm® QCS8450 | 147.799 ms | 8 - 2739 MB | NPU
80
+ | proposal_generator | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 122.505 ms | 7 - 71 MB | NPU
81
+ | proposal_generator | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 92.038 ms | 7 - 7 MB | NPU
82
+ | proposal_generator | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 53.247 ms | 0 - 1510 MB | NPU
83
+ | proposal_generator | QNN_DLC | float | Qualcomm® SA7255P | 353.632 ms | 2 - 1759 MB | NPU
84
+ | proposal_generator | QNN_DLC | float | Qualcomm® SA8295P | 125.972 ms | 0 - 1430 MB | NPU
85
+ | proposal_generator | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 53.247 ms | 0 - 1510 MB | NPU
86
+ | proposal_generator | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 42.128 ms | 7 - 1789 MB | NPU
87
+ | roi_head | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 72.432 ms | 244 - 718 MB | NPU
88
+ | roi_head | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 138.03 ms | 234 - 680 MB | NPU
89
+ | roi_head | ONNX | float | Qualcomm® Dragonwing™ IQ-8275 | 154.653 ms | 108 - 208 MB | NPU
90
+ | roi_head | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 87.331 ms | 271 - 273 MB | NPU
91
+ | roi_head | ONNX | float | Qualcomm® QCS8450 | 138.03 ms | 234 - 680 MB | NPU
92
+ | roi_head | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 109.801 ms | 112 - 211 MB | NPU
93
+ | roi_head | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 62.278 ms | 226 - 623 MB | NPU
94
+ | roi_head | ONNX | float | Snapdragon® 8 Elite Mobile | 62.278 ms | 226 - 623 MB | NPU
95
+ | roi_head | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 50.521 ms | 190 - 588 MB | NPU
96
+ | roi_head | QNN_DLC | float | Snapdragon® X2 Elite | 89.841 ms | 52 - 52 MB | NPU
97
+ | roi_head | QNN_DLC | float | Snapdragon® X Elite | 231.039 ms | 52 - 52 MB | NPU
98
+ | roi_head | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 159.1 ms | 10 - 795 MB | NPU
99
+ | roi_head | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 295.243 ms | 39 - 787 MB | NPU
100
+ | roi_head | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 243.435 ms | 52 - 107 MB | NPU
101
+ | roi_head | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 224.388 ms | 51 - 52 MB | NPU
102
+ | roi_head | QNN_DLC | float | Qualcomm® SA8775P | 261.792 ms | 49 - 777 MB | NPU
103
+ | roi_head | QNN_DLC | float | Qualcomm® SA8650P | 261.792 ms | 49 - 777 MB | NPU
104
+ | roi_head | QNN_DLC | float | Qualcomm® SA8255P | 261.792 ms | 49 - 777 MB | NPU
105
+ | roi_head | QNN_DLC | float | Qualcomm® QCS8450 | 295.243 ms | 39 - 787 MB | NPU
106
+ | roi_head | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 256.185 ms | 52 - 106 MB | NPU
107
+ | roi_head | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 231.039 ms | 52 - 52 MB | NPU
108
+ | roi_head | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 116.843 ms | 1 - 490 MB | NPU
109
+ | roi_head | QNN_DLC | float | Qualcomm® SA7255P | 560.942 ms | 31 - 534 MB | NPU
110
+ | roi_head | QNN_DLC | float | Qualcomm® SA8295P | 287.976 ms | 49 - 769 MB | NPU
111
+ | roi_head | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 116.843 ms | 1 - 490 MB | NPU
112
+ | roi_head | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 82.518 ms | 13 - 513 MB | NPU
 
 
113
 
114
  ## License
115
  * The license for the original implementation of MaskRCNN can be found
release_assets.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "version": "0.61.0",
3
  "precisions": {
4
  "float": {
5
  "universal_assets": {
@@ -8,13 +8,13 @@
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/maskrcnn/releases/v0.61.0/maskrcnn-onnx-float.zip"
12
  },
13
  "qnn_dlc": {
14
  "tool_versions": {
15
  "qairt": "2.45.0.260326154327"
16
  },
17
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/maskrcnn/releases/v0.61.0/maskrcnn-qnn_dlc-float.zip"
18
  }
19
  }
20
  }
 
1
  {
2
+ "version": "0.62.0",
3
  "precisions": {
4
  "float": {
5
  "universal_assets": {
 
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/maskrcnn/releases/v0.62.0/maskrcnn-onnx-float.zip"
12
  },
13
  "qnn_dlc": {
14
  "tool_versions": {
15
  "qairt": "2.45.0.260326154327"
16
  },
17
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/maskrcnn/releases/v0.62.0/maskrcnn-qnn_dlc-float.zip"
18
  }
19
  }
20
  }