Hugging Face's logo

qualcomm
/
FCN-ResNet50

Image Segmentation
PyTorch
android
Model card Files
xet
 Community
qaihm-bot commited on
Commit
e1cdef1
·
verified ·
1 Parent(s): 9e85c30

v0.60.0

Browse files

See https://github.com/qualcomm/ai-hub-models/releases/v0.60.0 for changelog.

Files changed (2) hide show
  1. README.md +114 -104
  2. release_assets.json +7 -7
README.md CHANGED
@@ -14,7 +14,7 @@ pipeline_tag: image-segmentation
14
  FCN_ResNet50 is a machine learning model that can segment images from the COCO dataset. It uses ResNet50 as a backbone.
15
 
16
  This is based on the implementation of FCN-ResNet50 found [here](https://github.com/pytorch/vision/blob/main/torchvision/models/segmentation/fcn.py).
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/v0.59.0/src/qai_hub_models/models/fcn_resnet50) 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
 
@@ -27,134 +27,144 @@ 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/fcn_resnet50/releases/v0.59.0/fcn_resnet50-onnx-float.zip)
31
- | ONNX | w8a8 | Universal | QAIRT 2.45, ONNX Runtime 1.27.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/fcn_resnet50/releases/v0.59.0/fcn_resnet50-onnx-w8a8.zip)
32
- | QNN_DLC | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/fcn_resnet50/releases/v0.59.0/fcn_resnet50-qnn_dlc-float.zip)
33
- | QNN_DLC | w8a8 | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/fcn_resnet50/releases/v0.59.0/fcn_resnet50-qnn_dlc-w8a8.zip)
34
- | TFLITE | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/fcn_resnet50/releases/v0.59.0/fcn_resnet50-tflite-float.zip)
35
- | TFLITE | w8a8 | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/fcn_resnet50/releases/v0.59.0/fcn_resnet50-tflite-w8a8.zip)
36
 
37
  For more device-specific assets and performance metrics, visit **[FCN-ResNet50 on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/fcn_resnet50)**.
38
 
39
 
40
  ### Option 2: Export with Custom Configurations
41
 
42
- Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/v0.59.0/src/qai_hub_models/models/fcn_resnet50) Python library to compile and export the model with your own:
43
  - Custom weights (e.g., fine-tuned checkpoints)
44
  - Custom input shapes
45
  - Target device and runtime configurations
46
 
47
  This option is ideal if you need to customize the model beyond the default configuration provided here.
48
 
49
- See our repository for [FCN-ResNet50 on GitHub](https://github.com/qualcomm/ai-hub-models/blob/v0.59.0/src/qai_hub_models/models/fcn_resnet50) for usage instructions.
50
 
51
  ## Model Details
52
 
53
  **Model Type:** Model_use_case.semantic_segmentation
54
 
55
  **Model Stats:**
56
- - Model checkpoint: COCO_WITH_VOC_LABELS_V1
57
  - Input resolution: 224x224
58
- - Number of output classes: 21
59
- - Number of parameters: 33.0M
60
  - Model size (float): 126 MB
61
  - Model size (w8a8): 32.2 MB
 
 
62
 
63
  ## Performance Summary
64
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
65
  |---|---|---|---|---|---|---
66
- | FCN-ResNet50 | ONNX | float | Snapdragon® X2 Elite | 22.244 ms | 3 - 3 MB | NPU
67
- | FCN-ResNet50 | ONNX | float | Snapdragon® X Elite | 43.629 ms | 64 - 64 MB | NPU
68
- | FCN-ResNet50 | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 32.337 ms | 1 - 381 MB | NPU
69
- | FCN-ResNet50 | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 83.182 ms | 3 - 273 MB | NPU
70
- | FCN-ResNet50 | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 43.888 ms | 3 - 7 MB | NPU
71
- | FCN-ResNet50 | ONNX | float | Qualcomm® QCS8450 | 83.182 ms | 3 - 273 MB | NPU
72
- | FCN-ResNet50 | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 72.739 ms | 3 - 9 MB | NPU
73
- | FCN-ResNet50 | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 43.629 ms | 64 - 64 MB | NPU
74
- | FCN-ResNet50 | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 26.55 ms | 2 - 311 MB | NPU
75
- | FCN-ResNet50 | ONNX | float | Snapdragon® 8 Elite Mobile | 26.55 ms | 2 - 311 MB | NPU
76
- | FCN-ResNet50 | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 21.089 ms | 2 - 329 MB | NPU
77
- | FCN-ResNet50 | ONNX | w8a8 | Snapdragon® X Elite | 13.785 ms | 32 - 32 MB | NPU
78
- | FCN-ResNet50 | ONNX | w8a8 | Snapdragon® 8 Gen 3 Mobile | 10.055 ms | 0 - 277 MB | NPU
79
- | FCN-ResNet50 | ONNX | w8a8 | Snapdragon® 8 Gen 1 Mobile | 21.004 ms | 1 - 275 MB | NPU
80
- | FCN-ResNet50 | ONNX | w8a8 | Qualcomm® Dragonwing™ QCS6490 | 121.804 ms | 1 - 4 MB | NPU
81
- | FCN-ResNet50 | ONNX | w8a8 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 13.598 ms | 0 - 51 MB | NPU
82
- | FCN-ResNet50 | ONNX | w8a8 | Qualcomm® QCS8450 | 21.004 ms | 1 - 275 MB | NPU
83
- | FCN-ResNet50 | ONNX | w8a8 | Qualcomm® Dragonwing™ IQ-9075 | 14.256 ms | 1 - 4 MB | NPU
84
- | FCN-ResNet50 | ONNX | w8a8 | Qualcomm® Dragonwing™ IQ-X7181 | 13.785 ms | 32 - 32 MB | NPU
85
- | FCN-ResNet50 | ONNX | w8a8 | Qualcomm® Dragonwing™ Q-8750 | 8.765 ms | 1 - 207 MB | NPU
86
- | FCN-ResNet50 | ONNX | w8a8 | Snapdragon® 8 Elite Mobile | 8.765 ms | 1 - 207 MB | NPU
87
- | FCN-ResNet50 | ONNX | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 7.424 ms | 0 - 251 MB | NPU
88
- | FCN-ResNet50 | QNN_DLC | float | Snapdragon® X2 Elite | 22.746 ms | 3 - 3 MB | NPU
89
- | FCN-ResNet50 | QNN_DLC | float | Snapdragon® X Elite | 44.81 ms | 3 - 3 MB | NPU
90
- | FCN-ResNet50 | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 33.162 ms | 3 - 378 MB | NPU
91
- | FCN-ResNet50 | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 83.741 ms | 3 - 270 MB | NPU
92
- | FCN-ResNet50 | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8275 | 273.681 ms | 1 - 309 MB | NPU
93
- | FCN-ResNet50 | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 44.166 ms | 3 - 6 MB | NPU
94
- | FCN-ResNet50 | QNN_DLC | float | Qualcomm® SA8775P | 72.857 ms | 1 - 308 MB | NPU
95
- | FCN-ResNet50 | QNN_DLC | float | Qualcomm® SA8650P | 72.857 ms | 1 - 308 MB | NPU
96
- | FCN-ResNet50 | QNN_DLC | float | Qualcomm® SA8255P | 72.857 ms | 1 - 308 MB | NPU
97
- | FCN-ResNet50 | QNN_DLC | float | Qualcomm® QCS8450 | 83.741 ms | 3 - 270 MB | NPU
98
- | FCN-ResNet50 | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 73.481 ms | 3 - 8 MB | NPU
99
- | FCN-ResNet50 | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 44.81 ms | 3 - 3 MB | NPU
100
- | FCN-ResNet50 | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 26.481 ms | 3 - 323 MB | NPU
101
- | FCN-ResNet50 | QNN_DLC | float | Qualcomm® SA7255P | 273.681 ms | 1 - 309 MB | NPU
102
- | FCN-ResNet50 | QNN_DLC | float | Qualcomm® SA8295P | 77.789 ms | 0 - 216 MB | NPU
103
- | FCN-ResNet50 | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 26.481 ms | 3 - 323 MB | NPU
104
- | FCN-ResNet50 | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 21.458 ms | 3 - 338 MB | NPU
105
- | FCN-ResNet50 | QNN_DLC | w8a8 | Snapdragon® X2 Elite | 7.924 ms | 1 - 1 MB | NPU
106
- | FCN-ResNet50 | QNN_DLC | w8a8 | Snapdragon® X Elite | 15.253 ms | 1 - 1 MB | NPU
107
- | FCN-ResNet50 | QNN_DLC | w8a8 | Snapdragon® 8 Gen 3 Mobile | 11.242 ms | 1 - 265 MB | NPU
108
- | FCN-ResNet50 | QNN_DLC | w8a8 | Snapdragon® 8 Gen 1 Mobile | 24.394 ms | 1 - 265 MB | NPU
109
- | FCN-ResNet50 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ QCS6490 | 122.868 ms | 3 - 5 MB | NPU
110
- | FCN-ResNet50 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ QCS8275 | 39.474 ms | 1 - 207 MB | NPU
111
- | FCN-ResNet50 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 14.868 ms | 1 - 3 MB | NPU
112
- | FCN-ResNet50 | QNN_DLC | w8a8 | Qualcomm® SA8775P | 14.709 ms | 1 - 216 MB | NPU
113
- | FCN-ResNet50 | QNN_DLC | w8a8 | Qualcomm® SA8650P | 14.709 ms | 1 - 216 MB | NPU
114
- | FCN-ResNet50 | QNN_DLC | w8a8 | Qualcomm® SA8255P | 14.709 ms | 1 - 216 MB | NPU
115
- | FCN-ResNet50 | QNN_DLC | w8a8 | Qualcomm® QCS8450 | 24.394 ms | 1 - 265 MB | NPU
116
- | FCN-ResNet50 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ IQ-9075 | 14.91 ms | 1 - 3 MB | NPU
117
- | FCN-ResNet50 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ IQ-X7181 | 15.253 ms | 1 - 1 MB | NPU
118
- | FCN-ResNet50 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ Q-6690 | 380.909 ms | 1 - 382 MB | NPU
119
- | FCN-ResNet50 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ Q-7790 | 26.228 ms | 1 - 280 MB | NPU
120
- | FCN-ResNet50 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ Q-8750 | 8.933 ms | 1 - 200 MB | NPU
121
- | FCN-ResNet50 | QNN_DLC | w8a8 | Qualcomm® SA7255P | 39.474 ms | 1 - 207 MB | NPU
122
- | FCN-ResNet50 | QNN_DLC | w8a8 | Qualcomm® SA8295P | 21.917 ms | 1 - 209 MB | NPU
123
- | FCN-ResNet50 | QNN_DLC | w8a8 | Snapdragon® 8 Elite Mobile | 8.933 ms | 1 - 200 MB | NPU
124
- | FCN-ResNet50 | QNN_DLC | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 7.434 ms | 1 - 250 MB | NPU
125
- | FCN-ResNet50 | QNN_DLC | w8a8 | Snapdragon® 7 Gen 4 Mobile | 26.228 ms | 1 - 280 MB | NPU
126
- | FCN-ResNet50 | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 32.995 ms | 0 - 426 MB | NPU
127
- | FCN-ResNet50 | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 83.651 ms | 1 - 323 MB | NPU
128
- | FCN-ResNet50 | TFLITE | float | Qualcomm® Dragonwing™ QCS8275 | 273.505 ms | 1 - 354 MB | NPU
129
- | FCN-ResNet50 | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 44.318 ms | 0 - 3 MB | NPU
130
- | FCN-ResNet50 | TFLITE | float | Qualcomm® SA8775P | 72.838 ms | 1 - 355 MB | NPU
131
- | FCN-ResNet50 | TFLITE | float | Qualcomm® SA8650P | 72.838 ms | 1 - 355 MB | NPU
132
- | FCN-ResNet50 | TFLITE | float | Qualcomm® SA8255P | 72.838 ms | 1 - 355 MB | NPU
133
- | FCN-ResNet50 | TFLITE | float | Qualcomm® QCS8450 | 83.651 ms | 1 - 323 MB | NPU
134
- | FCN-ResNet50 | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 72.751 ms | 0 - 71 MB | NPU
135
- | FCN-ResNet50 | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 27.069 ms | 0 - 356 MB | NPU
136
- | FCN-ResNet50 | TFLITE | float | Qualcomm® SA7255P | 273.505 ms | 1 - 354 MB | NPU
137
- | FCN-ResNet50 | TFLITE | float | Qualcomm® SA8295P | 77.808 ms | 1 - 268 MB | NPU
138
- | FCN-ResNet50 | TFLITE | float | Snapdragon® 8 Elite Mobile | 27.069 ms | 0 - 356 MB | NPU
139
- | FCN-ResNet50 | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 20.985 ms | 0 - 367 MB | NPU
140
- | FCN-ResNet50 | TFLITE | w8a8 | Snapdragon® 8 Gen 3 Mobile | 10.977 ms | 0 - 266 MB | NPU
141
- | FCN-ResNet50 | TFLITE | w8a8 | Snapdragon® 8 Gen 1 Mobile | 23.203 ms | 0 - 267 MB | NPU
142
- | FCN-ResNet50 | TFLITE | w8a8 | Qualcomm® Dragonwing™ QCS6490 | 124.21 ms | 0 - 39 MB | NPU
143
- | FCN-ResNet50 | TFLITE | w8a8 | Qualcomm® Dragonwing™ QCS8275 | 38.23 ms | 0 - 207 MB | NPU
144
- | FCN-ResNet50 | TFLITE | w8a8 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 14.311 ms | 0 - 3 MB | NPU
145
- | FCN-ResNet50 | TFLITE | w8a8 | Qualcomm® SA8775P | 14.706 ms | 0 - 207 MB | NPU
146
- | FCN-ResNet50 | TFLITE | w8a8 | Qualcomm® SA8650P | 14.706 ms | 0 - 207 MB | NPU
147
- | FCN-ResNet50 | TFLITE | w8a8 | Qualcomm® SA8255P | 14.706 ms | 0 - 207 MB | NPU
148
- | FCN-ResNet50 | TFLITE | w8a8 | Qualcomm® QCS8450 | 23.203 ms | 0 - 267 MB | NPU
149
- | FCN-ResNet50 | TFLITE | w8a8 | Qualcomm® Dragonwing™ IQ-9075 | 14.995 ms | 0 - 36 MB | NPU
150
- | FCN-ResNet50 | TFLITE | w8a8 | Qualcomm® Dragonwing™ Q-6690 | 432.199 ms | 0 - 388 MB | NPU
151
- | FCN-ResNet50 | TFLITE | w8a8 | Qualcomm® Dragonwing™ Q-7790 | 28.114 ms | 0 - 280 MB | NPU
152
- | FCN-ResNet50 | TFLITE | w8a8 | Qualcomm® Dragonwing™ Q-8750 | 8.588 ms | 0 - 198 MB | NPU
153
- | FCN-ResNet50 | TFLITE | w8a8 | Qualcomm® SA7255P | 38.23 ms | 0 - 207 MB | NPU
154
- | FCN-ResNet50 | TFLITE | w8a8 | Qualcomm® SA8295P | 21.275 ms | 0 - 208 MB | NPU
155
- | FCN-ResNet50 | TFLITE | w8a8 | Snapdragon® 8 Elite Mobile | 8.588 ms | 0 - 198 MB | NPU
156
- | FCN-ResNet50 | TFLITE | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 7.407 ms | 0 - 244 MB | NPU
157
- | FCN-ResNet50 | TFLITE | w8a8 | Snapdragon® 7 Gen 4 Mobile | 28.114 ms | 0 - 280 MB | NPU
 
 
 
 
 
 
 
 
 
 
158
 
159
  ## License
160
  * The license for the original implementation of FCN-ResNet50 can be found
 
14
  FCN_ResNet50 is a machine learning model that can segment images from the COCO dataset. It uses ResNet50 as a backbone.
15
 
16
  This is based on the implementation of FCN-ResNet50 found [here](https://github.com/pytorch/vision/blob/main/torchvision/models/segmentation/fcn.py).
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/v0.60.0/src/qai_hub_models/models/fcn_resnet50) 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/fcn_resnet50/releases/v0.60.0/fcn_resnet50-onnx-float.zip)
31
+ | ONNX | w8a8 | Universal | QAIRT 2.45, ONNX Runtime 1.27.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/fcn_resnet50/releases/v0.60.0/fcn_resnet50-onnx-w8a8.zip)
32
+ | QNN_DLC | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/fcn_resnet50/releases/v0.60.0/fcn_resnet50-qnn_dlc-float.zip)
33
+ | QNN_DLC | w8a8 | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/fcn_resnet50/releases/v0.60.0/fcn_resnet50-qnn_dlc-w8a8.zip)
34
+ | TFLITE | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/fcn_resnet50/releases/v0.60.0/fcn_resnet50-tflite-float.zip)
35
+ | TFLITE | w8a8 | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/fcn_resnet50/releases/v0.60.0/fcn_resnet50-tflite-w8a8.zip)
36
 
37
  For more device-specific assets and performance metrics, visit **[FCN-ResNet50 on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/fcn_resnet50)**.
38
 
39
 
40
  ### Option 2: Export with Custom Configurations
41
 
42
+ Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/v0.60.0/src/qai_hub_models/models/fcn_resnet50) Python library to compile and export the model with your own:
43
  - Custom weights (e.g., fine-tuned checkpoints)
44
  - Custom input shapes
45
  - Target device and runtime configurations
46
 
47
  This option is ideal if you need to customize the model beyond the default configuration provided here.
48
 
49
+ See our repository for [FCN-ResNet50 on GitHub](https://github.com/qualcomm/ai-hub-models/blob/v0.60.0/src/qai_hub_models/models/fcn_resnet50) for usage instructions.
50
 
51
  ## Model Details
52
 
53
  **Model Type:** Model_use_case.semantic_segmentation
54
 
55
  **Model Stats:**
 
56
  - Input resolution: 224x224
57
+ - Model checkpoint: COCO_WITH_VOC_LABELS_V1
 
58
  - Model size (float): 126 MB
59
  - Model size (w8a8): 32.2 MB
60
+ - Number of output classes: 21
61
+ - Number of parameters: 33.0M
62
 
63
  ## Performance Summary
64
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
65
  |---|---|---|---|---|---|---
66
+ | FCN-ResNet50 | ONNX | float | Snapdragon® X2 Elite | 22.291 ms | 3 - 3 MB | NPU
67
+ | FCN-ResNet50 | ONNX | float | Snapdragon® X Elite | 43.58 ms | 63 - 63 MB | NPU
68
+ | FCN-ResNet50 | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 32.671 ms | 4 - 386 MB | NPU
69
+ | FCN-ResNet50 | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 82.935 ms | 2 - 272 MB | NPU
70
+ | FCN-ResNet50 | ONNX | float | Qualcomm® Dragonwing™ IQ-8275 | 82.312 ms | 3 - 10 MB | NPU
71
+ | FCN-ResNet50 | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 43.571 ms | 0 - 79 MB | NPU
72
+ | FCN-ResNet50 | ONNX | float | Qualcomm® QCS8450 | 82.935 ms | 2 - 272 MB | NPU
73
+ | FCN-ResNet50 | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 74.416 ms | 3 - 9 MB | NPU
74
+ | FCN-ResNet50 | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 43.58 ms | 63 - 63 MB | NPU
75
+ | FCN-ResNet50 | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 26.688 ms | 2 - 311 MB | NPU
76
+ | FCN-ResNet50 | ONNX | float | Snapdragon® 8 Elite Mobile | 26.688 ms | 2 - 311 MB | NPU
77
+ | FCN-ResNet50 | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 21.051 ms | 1 - 330 MB | NPU
78
+ | FCN-ResNet50 | ONNX | w8a8 | Snapdragon® X2 Elite | 7.247 ms | 2 - 2 MB | NPU
79
+ | FCN-ResNet50 | ONNX | w8a8 | Snapdragon® X Elite | 13.798 ms | 32 - 32 MB | NPU
80
+ | FCN-ResNet50 | ONNX | w8a8 | Snapdragon® 8 Gen 3 Mobile | 10.023 ms | 1 - 280 MB | NPU
81
+ | FCN-ResNet50 | ONNX | w8a8 | Snapdragon® 8 Gen 1 Mobile | 20.975 ms | 0 - 272 MB | NPU
82
+ | FCN-ResNet50 | ONNX | w8a8 | Qualcomm® Dragonwing™ QCS6490 | 122.126 ms | 1 - 4 MB | NPU
83
+ | FCN-ResNet50 | ONNX | w8a8 | Qualcomm® Dragonwing™ IQ-8275 | 14.373 ms | 1 - 5 MB | NPU
84
+ | FCN-ResNet50 | ONNX | w8a8 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 13.596 ms | 0 - 41 MB | NPU
85
+ | FCN-ResNet50 | ONNX | w8a8 | Qualcomm® QCS8450 | 20.975 ms | 0 - 272 MB | NPU
86
+ | FCN-ResNet50 | ONNX | w8a8 | Qualcomm® Dragonwing™ IQ-9075 | 14.214 ms | 1 - 4 MB | NPU
87
+ | FCN-ResNet50 | ONNX | w8a8 | Qualcomm® Dragonwing™ IQ-X7181 | 13.798 ms | 32 - 32 MB | NPU
88
+ | FCN-ResNet50 | ONNX | w8a8 | Qualcomm® Dragonwing™ Q-6690 | 433.69 ms | 1 - 384 MB | NPU
89
+ | FCN-ResNet50 | ONNX | w8a8 | Qualcomm® Dragonwing™ Q-7790 | 25.627 ms | 1 - 283 MB | NPU
90
+ | FCN-ResNet50 | ONNX | w8a8 | Qualcomm® Dragonwing™ Q-8750 | 8.749 ms | 1 - 206 MB | NPU
91
+ | FCN-ResNet50 | ONNX | w8a8 | Snapdragon® 8 Elite Mobile | 8.749 ms | 1 - 206 MB | NPU
92
+ | FCN-ResNet50 | ONNX | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 7.46 ms | 0 - 249 MB | NPU
93
+ | FCN-ResNet50 | ONNX | w8a8 | Snapdragon® 7 Gen 4 Mobile | 25.627 ms | 1 - 283 MB | NPU
94
+ | FCN-ResNet50 | QNN_DLC | float | Snapdragon® X2 Elite | 22.874 ms | 3 - 3 MB | NPU
95
+ | FCN-ResNet50 | QNN_DLC | float | Snapdragon® X Elite | 44.848 ms | 3 - 3 MB | NPU
96
+ | FCN-ResNet50 | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 33.272 ms | 3 - 382 MB | NPU
97
+ | FCN-ResNet50 | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 83.722 ms | 3 - 272 MB | NPU
98
+ | FCN-ResNet50 | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 82.303 ms | 3 - 9 MB | NPU
99
+ | FCN-ResNet50 | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 273.699 ms | 0 - 310 MB | NPU
100
+ | FCN-ResNet50 | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 44.571 ms | 3 - 6 MB | NPU
101
+ | FCN-ResNet50 | QNN_DLC | float | Qualcomm® SA8775P | 72.879 ms | 1 - 310 MB | NPU
102
+ | FCN-ResNet50 | QNN_DLC | float | Qualcomm® SA8650P | 72.879 ms | 1 - 310 MB | NPU
103
+ | FCN-ResNet50 | QNN_DLC | float | Qualcomm® SA8255P | 72.879 ms | 1 - 310 MB | NPU
104
+ | FCN-ResNet50 | QNN_DLC | float | Qualcomm® QCS8450 | 83.722 ms | 3 - 272 MB | NPU
105
+ | FCN-ResNet50 | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 73.114 ms | 3 - 8 MB | NPU
106
+ | FCN-ResNet50 | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 44.848 ms | 3 - 3 MB | NPU
107
+ | FCN-ResNet50 | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 26.104 ms | 0 - 320 MB | NPU
108
+ | FCN-ResNet50 | QNN_DLC | float | Qualcomm® SA7255P | 273.699 ms | 0 - 310 MB | NPU
109
+ | FCN-ResNet50 | QNN_DLC | float | Qualcomm® SA8295P | 77.808 ms | 1 - 217 MB | NPU
110
+ | FCN-ResNet50 | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 26.104 ms | 0 - 320 MB | NPU
111
+ | FCN-ResNet50 | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 21.465 ms | 3 - 337 MB | NPU
112
+ | FCN-ResNet50 | QNN_DLC | w8a8 | Snapdragon® X2 Elite | 7.872 ms | 1 - 1 MB | NPU
113
+ | FCN-ResNet50 | QNN_DLC | w8a8 | Snapdragon® X Elite | 15.197 ms | 1 - 1 MB | NPU
114
+ | FCN-ResNet50 | QNN_DLC | w8a8 | Snapdragon® 8 Gen 3 Mobile | 11.219 ms | 1 - 263 MB | NPU
115
+ | FCN-ResNet50 | QNN_DLC | w8a8 | Snapdragon® 8 Gen 1 Mobile | 24.486 ms | 1 - 266 MB | NPU
116
+ | FCN-ResNet50 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ QCS6490 | 122.932 ms | 1 - 4 MB | NPU
117
+ | FCN-ResNet50 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ IQ-8275 | 15.832 ms | 1 - 4 MB | NPU
118
+ | FCN-ResNet50 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ IQ-8275 | 39.492 ms | 1 - 206 MB | NPU
119
+ | FCN-ResNet50 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 14.869 ms | 1 - 2 MB | NPU
120
+ | FCN-ResNet50 | QNN_DLC | w8a8 | Qualcomm® SA8775P | 14.736 ms | 1 - 215 MB | NPU
121
+ | FCN-ResNet50 | QNN_DLC | w8a8 | Qualcomm® SA8650P | 14.736 ms | 1 - 215 MB | NPU
122
+ | FCN-ResNet50 | QNN_DLC | w8a8 | Qualcomm® SA8255P | 14.736 ms | 1 - 215 MB | NPU
123
+ | FCN-ResNet50 | QNN_DLC | w8a8 | Qualcomm® QCS8450 | 24.486 ms | 1 - 266 MB | NPU
124
+ | FCN-ResNet50 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ IQ-9075 | 15.602 ms | 0 - 3 MB | NPU
125
+ | FCN-ResNet50 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ IQ-X7181 | 15.197 ms | 1 - 1 MB | NPU
126
+ | FCN-ResNet50 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ Q-6690 | 410.273 ms | 1 - 384 MB | NPU
127
+ | FCN-ResNet50 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ Q-7790 | 26.035 ms | 1 - 280 MB | NPU
128
+ | FCN-ResNet50 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ Q-8750 | 8.926 ms | 1 - 203 MB | NPU
129
+ | FCN-ResNet50 | QNN_DLC | w8a8 | Qualcomm® SA7255P | 39.492 ms | 1 - 206 MB | NPU
130
+ | FCN-ResNet50 | QNN_DLC | w8a8 | Qualcomm® SA8295P | 21.891 ms | 1 - 209 MB | NPU
131
+ | FCN-ResNet50 | QNN_DLC | w8a8 | Snapdragon® 8 Elite Mobile | 8.926 ms | 1 - 203 MB | NPU
132
+ | FCN-ResNet50 | QNN_DLC | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 7.469 ms | 1 - 250 MB | NPU
133
+ | FCN-ResNet50 | QNN_DLC | w8a8 | Snapdragon® 7 Gen 4 Mobile | 26.035 ms | 1 - 280 MB | NPU
134
+ | FCN-ResNet50 | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 33.289 ms | 22 - 452 MB | NPU
135
+ | FCN-ResNet50 | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 83.936 ms | 1 - 321 MB | NPU
136
+ | FCN-ResNet50 | TFLITE | float | Qualcomm® Dragonwing™ IQ-8275 | 82.105 ms | 0 - 72 MB | NPU
137
+ | FCN-ResNet50 | TFLITE | float | Qualcomm® Dragonwing™ IQ-8275 | 273.585 ms | 1 - 353 MB | NPU
138
+ | FCN-ResNet50 | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 44.557 ms | 0 - 87 MB | NPU
139
+ | FCN-ResNet50 | TFLITE | float | Qualcomm® SA8775P | 72.899 ms | 0 - 354 MB | NPU
140
+ | FCN-ResNet50 | TFLITE | float | Qualcomm® SA8650P | 72.899 ms | 0 - 354 MB | NPU
141
+ | FCN-ResNet50 | TFLITE | float | Qualcomm® SA8255P | 72.899 ms | 0 - 354 MB | NPU
142
+ | FCN-ResNet50 | TFLITE | float | Qualcomm® QCS8450 | 83.936 ms | 1 - 321 MB | NPU
143
+ | FCN-ResNet50 | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 73.02 ms | 0 - 71 MB | NPU
144
+ | FCN-ResNet50 | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 27.1 ms | 0 - 350 MB | NPU
145
+ | FCN-ResNet50 | TFLITE | float | Qualcomm® SA7255P | 273.585 ms | 1 - 353 MB | NPU
146
+ | FCN-ResNet50 | TFLITE | float | Qualcomm® SA8295P | 77.758 ms | 0 - 266 MB | NPU
147
+ | FCN-ResNet50 | TFLITE | float | Snapdragon® 8 Elite Mobile | 27.1 ms | 0 - 350 MB | NPU
148
+ | FCN-ResNet50 | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 20.929 ms | 0 - 372 MB | NPU
149
+ | FCN-ResNet50 | TFLITE | w8a8 | Snapdragon® 8 Gen 3 Mobile | 10.987 ms | 0 - 266 MB | NPU
150
+ | FCN-ResNet50 | TFLITE | w8a8 | Snapdragon® 8 Gen 1 Mobile | 23.148 ms | 0 - 267 MB | NPU
151
+ | FCN-ResNet50 | TFLITE | w8a8 | Qualcomm® Dragonwing™ QCS6490 | 124.314 ms | 0 - 39 MB | NPU
152
+ | FCN-ResNet50 | TFLITE | w8a8 | Qualcomm® Dragonwing™ IQ-8275 | 15.175 ms | 0 - 37 MB | NPU
153
+ | FCN-ResNet50 | TFLITE | w8a8 | Qualcomm® Dragonwing™ IQ-8275 | 38.226 ms | 0 - 206 MB | NPU
154
+ | FCN-ResNet50 | TFLITE | w8a8 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 14.272 ms | 0 - 2 MB | NPU
155
+ | FCN-ResNet50 | TFLITE | w8a8 | Qualcomm® SA8775P | 14.698 ms | 0 - 207 MB | NPU
156
+ | FCN-ResNet50 | TFLITE | w8a8 | Qualcomm® SA8650P | 14.698 ms | 0 - 207 MB | NPU
157
+ | FCN-ResNet50 | TFLITE | w8a8 | Qualcomm® SA8255P | 14.698 ms | 0 - 207 MB | NPU
158
+ | FCN-ResNet50 | TFLITE | w8a8 | Qualcomm® QCS8450 | 23.148 ms | 0 - 267 MB | NPU
159
+ | FCN-ResNet50 | TFLITE | w8a8 | Qualcomm® Dragonwing™ IQ-9075 | 15.067 ms | 0 - 36 MB | NPU
160
+ | FCN-ResNet50 | TFLITE | w8a8 | Qualcomm® Dragonwing™ Q-6690 | 428.846 ms | 0 - 389 MB | NPU
161
+ | FCN-ResNet50 | TFLITE | w8a8 | Qualcomm® Dragonwing™ Q-7790 | 27.887 ms | 0 - 280 MB | NPU
162
+ | FCN-ResNet50 | TFLITE | w8a8 | Qualcomm® Dragonwing™ Q-8750 | 8.577 ms | 0 - 198 MB | NPU
163
+ | FCN-ResNet50 | TFLITE | w8a8 | Qualcomm® SA7255P | 38.226 ms | 0 - 206 MB | NPU
164
+ | FCN-ResNet50 | TFLITE | w8a8 | Qualcomm® SA8295P | 21.306 ms | 0 - 211 MB | NPU
165
+ | FCN-ResNet50 | TFLITE | w8a8 | Snapdragon® 8 Elite Mobile | 8.577 ms | 0 - 198 MB | NPU
166
+ | FCN-ResNet50 | TFLITE | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 7.413 ms | 0 - 246 MB | NPU
167
+ | FCN-ResNet50 | TFLITE | w8a8 | Snapdragon® 7 Gen 4 Mobile | 27.887 ms | 0 - 280 MB | NPU
168
 
169
  ## License
170
  * The license for the original implementation of FCN-ResNet50 can be found
release_assets.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "version": "0.59.0",
3
  "precisions": {
4
  "float": {
5
  "universal_assets": {
@@ -8,19 +8,19 @@
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/fcn_resnet50/releases/v0.59.0/fcn_resnet50-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/fcn_resnet50/releases/v0.59.0/fcn_resnet50-qnn_dlc-float.zip"
18
  },
19
  "tflite": {
20
  "tool_versions": {
21
  "qairt": "2.45.0.260326154327"
22
  },
23
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/fcn_resnet50/releases/v0.59.0/fcn_resnet50-tflite-float.zip"
24
  }
25
  }
26
  },
@@ -31,19 +31,19 @@
31
  "qairt": "2.45.0.260326154327",
32
  "onnx_runtime": "1.27.1"
33
  },
34
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/fcn_resnet50/releases/v0.59.0/fcn_resnet50-onnx-w8a8.zip"
35
  },
36
  "qnn_dlc": {
37
  "tool_versions": {
38
  "qairt": "2.45.0.260326154327"
39
  },
40
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/fcn_resnet50/releases/v0.59.0/fcn_resnet50-qnn_dlc-w8a8.zip"
41
  },
42
  "tflite": {
43
  "tool_versions": {
44
  "qairt": "2.45.0.260326154327"
45
  },
46
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/fcn_resnet50/releases/v0.59.0/fcn_resnet50-tflite-w8a8.zip"
47
  }
48
  }
49
  }
 
1
  {
2
+ "version": "0.60.0",
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/fcn_resnet50/releases/v0.60.0/fcn_resnet50-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/fcn_resnet50/releases/v0.60.0/fcn_resnet50-qnn_dlc-float.zip"
18
  },
19
  "tflite": {
20
  "tool_versions": {
21
  "qairt": "2.45.0.260326154327"
22
  },
23
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/fcn_resnet50/releases/v0.60.0/fcn_resnet50-tflite-float.zip"
24
  }
25
  }
26
  },
 
31
  "qairt": "2.45.0.260326154327",
32
  "onnx_runtime": "1.27.1"
33
  },
34
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/fcn_resnet50/releases/v0.60.0/fcn_resnet50-onnx-w8a8.zip"
35
  },
36
  "qnn_dlc": {
37
  "tool_versions": {
38
  "qairt": "2.45.0.260326154327"
39
  },
40
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/fcn_resnet50/releases/v0.60.0/fcn_resnet50-qnn_dlc-w8a8.zip"
41
  },
42
  "tflite": {
43
  "tool_versions": {
44
  "qairt": "2.45.0.260326154327"
45
  },
46
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/fcn_resnet50/releases/v0.60.0/fcn_resnet50-tflite-w8a8.zip"
47
  }
48
  }
49
  }