Hugging Face's logo

qualcomm
/
DeepLabV3-Plus-MobileNet

Image Segmentation
PyTorch
android
Model card Files
xet
 Community
qaihm-bot commited on
Commit
d123cb9
·
verified ·
1 Parent(s): 990c9a9

v0.49.1

Browse files

See https://github.com/qualcomm/ai-hub-models/releases/v0.49.1 for changelog.

Files changed (1) hide show
  1. README.md +97 -97
README.md CHANGED
@@ -14,7 +14,7 @@ pipeline_tag: image-segmentation
14
  DeepLabV3 is designed for semantic segmentation at multiple scales, trained on the various datasets. It uses MobileNet as a backbone.
15
 
16
  This is based on the implementation of DeepLabV3-Plus-MobileNet found [here](https://github.com/jfzhang95/pytorch-deeplab-xception).
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/main/qai_hub_models/models/deeplabv3_plus_mobilenet) 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 +27,28 @@ Below are pre-exported model assets ready for deployment.
27
 
28
  | Runtime | Precision | Chipset | SDK Versions | Download |
29
  |---|---|---|---|---|
30
- | ONNX | float | Universal | QAIRT 2.42, ONNX Runtime 1.24.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/deeplabv3_plus_mobilenet/releases/v0.48.0/deeplabv3_plus_mobilenet-onnx-float.zip)
31
- | ONNX | w8a16 | Universal | QAIRT 2.42, ONNX Runtime 1.24.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/deeplabv3_plus_mobilenet/releases/v0.48.0/deeplabv3_plus_mobilenet-onnx-w8a16.zip)
32
- | ONNX | w8a8 | Universal | QAIRT 2.42, ONNX Runtime 1.24.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/deeplabv3_plus_mobilenet/releases/v0.48.0/deeplabv3_plus_mobilenet-onnx-w8a8.zip)
33
- | QNN_DLC | float | Universal | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/deeplabv3_plus_mobilenet/releases/v0.48.0/deeplabv3_plus_mobilenet-qnn_dlc-float.zip)
34
- | QNN_DLC | w8a16 | Universal | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/deeplabv3_plus_mobilenet/releases/v0.48.0/deeplabv3_plus_mobilenet-qnn_dlc-w8a16.zip)
35
- | QNN_DLC | w8a8 | Universal | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/deeplabv3_plus_mobilenet/releases/v0.48.0/deeplabv3_plus_mobilenet-qnn_dlc-w8a8.zip)
36
- | TFLITE | float | Universal | QAIRT 2.43, TFLite 2.17.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/deeplabv3_plus_mobilenet/releases/v0.48.0/deeplabv3_plus_mobilenet-tflite-float.zip)
37
- | TFLITE | w8a8 | Universal | QAIRT 2.43, TFLite 2.17.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/deeplabv3_plus_mobilenet/releases/v0.48.0/deeplabv3_plus_mobilenet-tflite-w8a8.zip)
38
 
39
  For more device-specific assets and performance metrics, visit **[DeepLabV3-Plus-MobileNet on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/deeplabv3_plus_mobilenet)**.
40
 
41
 
42
  ### Option 2: Export with Custom Configurations
43
 
44
- Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/main/qai_hub_models/models/deeplabv3_plus_mobilenet) Python library to compile and export the model with your own:
45
  - Custom weights (e.g., fine-tuned checkpoints)
46
  - Custom input shapes
47
  - Target device and runtime configurations
48
 
49
  This option is ideal if you need to customize the model beyond the default configuration provided here.
50
 
51
- See our repository for [DeepLabV3-Plus-MobileNet on GitHub](https://github.com/qualcomm/ai-hub-models/blob/main/qai_hub_models/models/deeplabv3_plus_mobilenet) for usage instructions.
52
 
53
  ## Model Details
54
 
@@ -65,95 +65,95 @@ See our repository for [DeepLabV3-Plus-MobileNet on GitHub](https://github.com/q
65
  ## Performance Summary
66
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
67
  |---|---|---|---|---|---|---
68
- | DeepLabV3-Plus-MobileNet | ONNX | float | Snapdragon® X2 Elite | 5.383 ms | 10 - 10 MB | NPU
69
- | DeepLabV3-Plus-MobileNet | ONNX | float | Snapdragon® X Elite | 11.032 ms | 10 - 10 MB | NPU
70
- | DeepLabV3-Plus-MobileNet | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 7.204 ms | 4 - 215 MB | NPU
71
- | DeepLabV3-Plus-MobileNet | ONNX | float | Qualcomm® QCS8550 (Proxy) | 10.345 ms | 3 - 5 MB | NPU
72
- | DeepLabV3-Plus-MobileNet | ONNX | float | Qualcomm® QCS9075 | 18.027 ms | 3 - 6 MB | NPU
73
- | DeepLabV3-Plus-MobileNet | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 5.894 ms | 1 - 170 MB | NPU
74
- | DeepLabV3-Plus-MobileNet | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 4.303 ms | 1 - 175 MB | NPU
75
- | DeepLabV3-Plus-MobileNet | ONNX | w8a16 | Snapdragon® X2 Elite | 3.716 ms | 7 - 7 MB | NPU
76
- | DeepLabV3-Plus-MobileNet | ONNX | w8a16 | Snapdragon® X Elite | 7.688 ms | 5 - 5 MB | NPU
77
- | DeepLabV3-Plus-MobileNet | ONNX | w8a16 | Snapdragon® 8 Gen 3 Mobile | 5.097 ms | 2 - 219 MB | NPU
78
- | DeepLabV3-Plus-MobileNet | ONNX | w8a16 | Qualcomm® QCS6490 | 1221.485 ms | 89 - 92 MB | CPU
79
- | DeepLabV3-Plus-MobileNet | ONNX | w8a16 | Qualcomm® QCS8550 (Proxy) | 7.073 ms | 2 - 5 MB | NPU
80
- | DeepLabV3-Plus-MobileNet | ONNX | w8a16 | Qualcomm® QCS9075 | 8.616 ms | 2 - 4 MB | NPU
81
- | DeepLabV3-Plus-MobileNet | ONNX | w8a16 | Qualcomm® QCM6690 | 589.325 ms | 107 - 115 MB | CPU
82
- | DeepLabV3-Plus-MobileNet | ONNX | w8a16 | Snapdragon® 8 Elite For Galaxy Mobile | 3.937 ms | 0 - 184 MB | NPU
83
- | DeepLabV3-Plus-MobileNet | ONNX | w8a16 | Snapdragon® 7 Gen 4 Mobile | 585.899 ms | 92 - 101 MB | CPU
84
- | DeepLabV3-Plus-MobileNet | ONNX | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 2.962 ms | 0 - 210 MB | NPU
85
- | DeepLabV3-Plus-MobileNet | ONNX | w8a8 | Snapdragon® X2 Elite | 1.672 ms | 7 - 7 MB | NPU
86
- | DeepLabV3-Plus-MobileNet | ONNX | w8a8 | Snapdragon® X Elite | 3.927 ms | 5 - 5 MB | NPU
87
- | DeepLabV3-Plus-MobileNet | ONNX | w8a8 | Snapdragon® 8 Gen 3 Mobile | 2.374 ms | 0 - 203 MB | NPU
88
- | DeepLabV3-Plus-MobileNet | ONNX | w8a8 | Qualcomm® QCS8550 (Proxy) | 3.537 ms | 0 - 3 MB | NPU
89
- | DeepLabV3-Plus-MobileNet | ONNX | w8a8 | Qualcomm® QCS9075 | 4.461 ms | 1 - 4 MB | NPU
90
- | DeepLabV3-Plus-MobileNet | ONNX | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 1.904 ms | 0 - 176 MB | NPU
91
- | DeepLabV3-Plus-MobileNet | ONNX | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 1.462 ms | 0 - 179 MB | NPU
92
- | DeepLabV3-Plus-MobileNet | QNN_DLC | float | Snapdragon® X2 Elite | 5.613 ms | 3 - 3 MB | NPU
93
- | DeepLabV3-Plus-MobileNet | QNN_DLC | float | Snapdragon® X Elite | 12.228 ms | 3 - 3 MB | NPU
94
- | DeepLabV3-Plus-MobileNet | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 8.047 ms | 3 - 213 MB | NPU
95
- | DeepLabV3-Plus-MobileNet | QNN_DLC | float | Qualcomm® QCS8275 (Proxy) | 58.591 ms | 1 - 167 MB | NPU
96
- | DeepLabV3-Plus-MobileNet | QNN_DLC | float | Qualcomm® QCS8550 (Proxy) | 11.508 ms | 3 - 5 MB | NPU
97
- | DeepLabV3-Plus-MobileNet | QNN_DLC | float | Qualcomm® SA8775P | 17.521 ms | 0 - 171 MB | NPU
98
- | DeepLabV3-Plus-MobileNet | QNN_DLC | float | Qualcomm® QCS9075 | 20.252 ms | 5 - 10 MB | NPU
99
- | DeepLabV3-Plus-MobileNet | QNN_DLC | float | Qualcomm® QCS8450 (Proxy) | 18.981 ms | 1 - 211 MB | NPU
100
- | DeepLabV3-Plus-MobileNet | QNN_DLC | float | Qualcomm® SA7255P | 58.591 ms | 1 - 167 MB | NPU
101
- | DeepLabV3-Plus-MobileNet | QNN_DLC | float | Qualcomm® SA8295P | 19.058 ms | 0 - 174 MB | NPU
102
- | DeepLabV3-Plus-MobileNet | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 6.25 ms | 3 - 199 MB | NPU
103
- | DeepLabV3-Plus-MobileNet | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 4.295 ms | 3 - 195 MB | NPU
104
- | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a16 | Snapdragon® X2 Elite | 4.163 ms | 2 - 2 MB | NPU
105
- | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a16 | Snapdragon® X Elite | 9.147 ms | 2 - 2 MB | NPU
106
- | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a16 | Snapdragon® 8 Gen 3 Mobile | 6.112 ms | 0 - 218 MB | NPU
107
- | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a16 | Qualcomm® QCS6490 | 33.548 ms | 1 - 4 MB | NPU
 
108
  | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a16 | Qualcomm® QCS8275 (Proxy) | 21.941 ms | 2 - 182 MB | NPU
109
- | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a16 | Qualcomm® QCS8550 (Proxy) | 8.473 ms | 2 - 4 MB | NPU
110
- | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a16 | Qualcomm® SA8775P | 40.099 ms | 2 - 181 MB | NPU
111
- | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a16 | Qualcomm® QCS9075 | 9.797 ms | 1 - 4 MB | NPU
112
- | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a16 | Qualcomm® QCM6690 | 100.543 ms | 2 - 233 MB | NPU
113
- | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a16 | Qualcomm® QCS8450 (Proxy) | 12.062 ms | 2 - 222 MB | NPU
114
  | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a16 | Qualcomm® SA7255P | 21.941 ms | 2 - 182 MB | NPU
115
- | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a16 | Qualcomm® SA8295P | 13.74 ms | 2 - 197 MB | NPU
116
- | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a16 | Snapdragon® 8 Elite For Galaxy Mobile | 4.328 ms | 2 - 181 MB | NPU
117
- | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a16 | Snapdragon® 7 Gen 4 Mobile | 12.099 ms | 2 - 180 MB | NPU
118
- | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 3.073 ms | 2 - 197 MB | NPU
119
- | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a8 | Snapdragon® X2 Elite | 2.055 ms | 1 - 1 MB | NPU
120
- | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a8 | Snapdragon® X Elite | 4.533 ms | 1 - 1 MB | NPU
121
- | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a8 | Snapdragon® 8 Gen 3 Mobile | 2.9 ms | 0 - 196 MB | NPU
122
- | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a8 | Qualcomm® QCS6490 | 15.752 ms | 3 - 5 MB | NPU
123
- | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a8 | Qualcomm® QCS8275 (Proxy) | 10.789 ms | 1 - 170 MB | NPU
124
- | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a8 | Qualcomm® QCS8550 (Proxy) | 4.123 ms | 1 - 2 MB | NPU
125
- | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a8 | Qualcomm® SA8775P | 4.62 ms | 1 - 172 MB | NPU
126
- | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a8 | Qualcomm® QCS9075 | 5.015 ms | 1 - 3 MB | NPU
127
- | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a8 | Qualcomm® QCM6690 | 47.555 ms | 1 - 202 MB | NPU
128
- | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a8 | Qualcomm® QCS8450 (Proxy) | 6.421 ms | 1 - 198 MB | NPU
129
- | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a8 | Qualcomm® SA7255P | 10.789 ms | 1 - 170 MB | NPU
130
- | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a8 | Qualcomm® SA8295P | 6.412 ms | 1 - 167 MB | NPU
131
- | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 2.14 ms | 1 - 171 MB | NPU
132
- | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a8 | Snapdragon® 7 Gen 4 Mobile | 6.2 ms | 1 - 175 MB | NPU
133
  | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 1.507 ms | 1 - 175 MB | NPU
134
- | DeepLabV3-Plus-MobileNet | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 8.052 ms | 0 - 221 MB | NPU
135
- | DeepLabV3-Plus-MobileNet | TFLITE | float | Qualcomm® QCS8275 (Proxy) | 58.6 ms | 0 - 170 MB | NPU
136
- | DeepLabV3-Plus-MobileNet | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 11.505 ms | 0 - 3 MB | NPU
137
- | DeepLabV3-Plus-MobileNet | TFLITE | float | Qualcomm® SA8775P | 17.571 ms | 0 - 174 MB | NPU
138
- | DeepLabV3-Plus-MobileNet | TFLITE | float | Qualcomm® QCS9075 | 19.652 ms | 0 - 18 MB | NPU
139
- | DeepLabV3-Plus-MobileNet | TFLITE | float | Qualcomm® QCS8450 (Proxy) | 18.977 ms | 0 - 217 MB | NPU
140
- | DeepLabV3-Plus-MobileNet | TFLITE | float | Qualcomm® SA7255P | 58.6 ms | 0 - 170 MB | NPU
141
- | DeepLabV3-Plus-MobileNet | TFLITE | float | Qualcomm® SA8295P | 19.131 ms | 0 - 174 MB | NPU
142
- | DeepLabV3-Plus-MobileNet | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 6.244 ms | 0 - 199 MB | NPU
143
- | DeepLabV3-Plus-MobileNet | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 4.294 ms | 0 - 193 MB | NPU
144
- | DeepLabV3-Plus-MobileNet | TFLITE | w8a8 | Snapdragon® 8 Gen 3 Mobile | 2.955 ms | 0 - 207 MB | NPU
145
- | DeepLabV3-Plus-MobileNet | TFLITE | w8a8 | Qualcomm® QCS6490 | 15.843 ms | 0 - 11 MB | NPU
146
- | DeepLabV3-Plus-MobileNet | TFLITE | w8a8 | Qualcomm® QCS8275 (Proxy) | 10.875 ms | 0 - 173 MB | NPU
147
- | DeepLabV3-Plus-MobileNet | TFLITE | w8a8 | Qualcomm® QCS8550 (Proxy) | 4.31 ms | 0 - 10 MB | NPU
148
- | DeepLabV3-Plus-MobileNet | TFLITE | w8a8 | Qualcomm® SA8775P | 4.658 ms | 0 - 175 MB | NPU
149
- | DeepLabV3-Plus-MobileNet | TFLITE | w8a8 | Qualcomm® QCS9075 | 5.166 ms | 0 - 10 MB | NPU
150
- | DeepLabV3-Plus-MobileNet | TFLITE | w8a8 | Qualcomm® QCM6690 | 52.604 ms | 0 - 193 MB | NPU
151
- | DeepLabV3-Plus-MobileNet | TFLITE | w8a8 | Qualcomm® QCS8450 (Proxy) | 5.75 ms | 0 - 205 MB | NPU
152
- | DeepLabV3-Plus-MobileNet | TFLITE | w8a8 | Qualcomm® SA7255P | 10.875 ms | 0 - 173 MB | NPU
153
- | DeepLabV3-Plus-MobileNet | TFLITE | w8a8 | Qualcomm® SA8295P | 6.241 ms | 0 - 172 MB | NPU
154
- | DeepLabV3-Plus-MobileNet | TFLITE | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 2.093 ms | 0 - 177 MB | NPU
155
- | DeepLabV3-Plus-MobileNet | TFLITE | w8a8 | Snapdragon® 7 Gen 4 Mobile | 6.365 ms | 0 - 178 MB | NPU
156
- | DeepLabV3-Plus-MobileNet | TFLITE | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 1.523 ms | 0 - 182 MB | NPU
 
 
 
 
 
 
 
 
 
 
 
 
 
 
157
 
158
  ## License
159
  * The license for the original implementation of DeepLabV3-Plus-MobileNet can be found
 
14
  DeepLabV3 is designed for semantic segmentation at multiple scales, trained on the various datasets. It uses MobileNet as a backbone.
15
 
16
  This is based on the implementation of DeepLabV3-Plus-MobileNet found [here](https://github.com/jfzhang95/pytorch-deeplab-xception).
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/tree/v0.49.1/qai_hub_models/models/deeplabv3_plus_mobilenet) 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 2.42, ONNX Runtime 1.24.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/deeplabv3_plus_mobilenet/releases/v0.49.1/deeplabv3_plus_mobilenet-onnx-float.zip)
31
+ | ONNX | w8a16 | Universal | QAIRT 2.42, ONNX Runtime 1.24.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/deeplabv3_plus_mobilenet/releases/v0.49.1/deeplabv3_plus_mobilenet-onnx-w8a16.zip)
32
+ | ONNX | w8a8 | Universal | QAIRT 2.42, ONNX Runtime 1.24.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/deeplabv3_plus_mobilenet/releases/v0.49.1/deeplabv3_plus_mobilenet-onnx-w8a8.zip)
33
+ | QNN_DLC | float | Universal | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/deeplabv3_plus_mobilenet/releases/v0.49.1/deeplabv3_plus_mobilenet-qnn_dlc-float.zip)
34
+ | QNN_DLC | w8a16 | Universal | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/deeplabv3_plus_mobilenet/releases/v0.49.1/deeplabv3_plus_mobilenet-qnn_dlc-w8a16.zip)
35
+ | QNN_DLC | w8a8 | Universal | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/deeplabv3_plus_mobilenet/releases/v0.49.1/deeplabv3_plus_mobilenet-qnn_dlc-w8a8.zip)
36
+ | TFLITE | float | Universal | QAIRT 2.43, TFLite 2.17.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/deeplabv3_plus_mobilenet/releases/v0.49.1/deeplabv3_plus_mobilenet-tflite-float.zip)
37
+ | TFLITE | w8a8 | Universal | QAIRT 2.43, TFLite 2.17.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/deeplabv3_plus_mobilenet/releases/v0.49.1/deeplabv3_plus_mobilenet-tflite-w8a8.zip)
38
 
39
  For more device-specific assets and performance metrics, visit **[DeepLabV3-Plus-MobileNet on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/deeplabv3_plus_mobilenet)**.
40
 
41
 
42
  ### Option 2: Export with Custom Configurations
43
 
44
+ Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/tree/v0.49.1/qai_hub_models/models/deeplabv3_plus_mobilenet) Python library to compile and export the model with your own:
45
  - Custom weights (e.g., fine-tuned checkpoints)
46
  - Custom input shapes
47
  - Target device and runtime configurations
48
 
49
  This option is ideal if you need to customize the model beyond the default configuration provided here.
50
 
51
+ See our repository for [DeepLabV3-Plus-MobileNet on GitHub](https://github.com/qualcomm/ai-hub-models/tree/v0.49.1/qai_hub_models/models/deeplabv3_plus_mobilenet) for usage instructions.
52
 
53
  ## Model Details
54
 
 
65
  ## Performance Summary
66
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
67
  |---|---|---|---|---|---|---
68
+ | DeepLabV3-Plus-MobileNet | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 4.301 ms | 4 - 177 MB | NPU
69
+ | DeepLabV3-Plus-MobileNet | ONNX | float | Snapdragon® X2 Elite | 5.394 ms | 10 - 10 MB | NPU
70
+ | DeepLabV3-Plus-MobileNet | ONNX | float | Snapdragon® X Elite | 11.015 ms | 10 - 10 MB | NPU
71
+ | DeepLabV3-Plus-MobileNet | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 7.196 ms | 4 - 214 MB | NPU
72
+ | DeepLabV3-Plus-MobileNet | ONNX | float | Qualcomm® QCS8550 (Proxy) | 10.348 ms | 3 - 40 MB | NPU
73
+ | DeepLabV3-Plus-MobileNet | ONNX | float | Qualcomm® QCS9075 | 18.064 ms | 3 - 6 MB | NPU
74
+ | DeepLabV3-Plus-MobileNet | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 5.888 ms | 1 - 169 MB | NPU
75
+ | DeepLabV3-Plus-MobileNet | ONNX | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 2.967 ms | 0 - 207 MB | NPU
76
+ | DeepLabV3-Plus-MobileNet | ONNX | w8a16 | Snapdragon® X2 Elite | 3.713 ms | 7 - 7 MB | NPU
77
+ | DeepLabV3-Plus-MobileNet | ONNX | w8a16 | Snapdragon® X Elite | 7.647 ms | 5 - 5 MB | NPU
78
+ | DeepLabV3-Plus-MobileNet | ONNX | w8a16 | Snapdragon® 8 Gen 3 Mobile | 5.092 ms | 2 - 219 MB | NPU
79
+ | DeepLabV3-Plus-MobileNet | ONNX | w8a16 | Qualcomm® QCS6490 | 1221.534 ms | 89 - 93 MB | CPU
80
+ | DeepLabV3-Plus-MobileNet | ONNX | w8a16 | Qualcomm® QCS8550 (Proxy) | 7.067 ms | 2 - 4 MB | NPU
81
+ | DeepLabV3-Plus-MobileNet | ONNX | w8a16 | Qualcomm® QCS9075 | 8.607 ms | 2 - 4 MB | NPU
82
+ | DeepLabV3-Plus-MobileNet | ONNX | w8a16 | Qualcomm® QCM6690 | 591.029 ms | 103 - 111 MB | CPU
83
+ | DeepLabV3-Plus-MobileNet | ONNX | w8a16 | Snapdragon® 8 Elite For Galaxy Mobile | 3.943 ms | 0 - 184 MB | NPU
84
+ | DeepLabV3-Plus-MobileNet | ONNX | w8a16 | Snapdragon® 7 Gen 4 Mobile | 585.572 ms | 94 - 102 MB | CPU
85
+ | DeepLabV3-Plus-MobileNet | ONNX | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 1.463 ms | 0 - 179 MB | NPU
86
+ | DeepLabV3-Plus-MobileNet | ONNX | w8a8 | Snapdragon® X2 Elite | 1.67 ms | 7 - 7 MB | NPU
87
+ | DeepLabV3-Plus-MobileNet | ONNX | w8a8 | Snapdragon® X Elite | 3.909 ms | 5 - 5 MB | NPU
88
+ | DeepLabV3-Plus-MobileNet | ONNX | w8a8 | Snapdragon® 8 Gen 3 Mobile | 2.377 ms | 0 - 205 MB | NPU
89
+ | DeepLabV3-Plus-MobileNet | ONNX | w8a8 | Qualcomm® QCS8550 (Proxy) | 3.504 ms | 0 - 2 MB | NPU
90
+ | DeepLabV3-Plus-MobileNet | ONNX | w8a8 | Qualcomm® QCS9075 | 4.473 ms | 1 - 4 MB | NPU
91
+ | DeepLabV3-Plus-MobileNet | ONNX | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 1.903 ms | 0 - 180 MB | NPU
92
+ | DeepLabV3-Plus-MobileNet | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 4.296 ms | 3 - 195 MB | NPU
93
+ | DeepLabV3-Plus-MobileNet | QNN_DLC | float | Snapdragon® X2 Elite | 5.604 ms | 3 - 3 MB | NPU
94
+ | DeepLabV3-Plus-MobileNet | QNN_DLC | float | Snapdragon® X Elite | 12.255 ms | 3 - 3 MB | NPU
95
+ | DeepLabV3-Plus-MobileNet | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 8.061 ms | 0 - 210 MB | NPU
96
+ | DeepLabV3-Plus-MobileNet | QNN_DLC | float | Qualcomm® QCS8275 (Proxy) | 58.594 ms | 1 - 167 MB | NPU
97
+ | DeepLabV3-Plus-MobileNet | QNN_DLC | float | Qualcomm® QCS8550 (Proxy) | 11.495 ms | 3 - 5 MB | NPU
98
+ | DeepLabV3-Plus-MobileNet | QNN_DLC | float | Qualcomm® SA8775P | 17.504 ms | 1 - 171 MB | NPU
99
+ | DeepLabV3-Plus-MobileNet | QNN_DLC | float | Qualcomm® QCS9075 | 20.158 ms | 5 - 10 MB | NPU
100
+ | DeepLabV3-Plus-MobileNet | QNN_DLC | float | Qualcomm® QCS8450 (Proxy) | 19.033 ms | 3 - 213 MB | NPU
101
+ | DeepLabV3-Plus-MobileNet | QNN_DLC | float | Qualcomm® SA7255P | 58.594 ms | 1 - 167 MB | NPU
102
+ | DeepLabV3-Plus-MobileNet | QNN_DLC | float | Qualcomm® SA8295P | 19.052 ms | 0 - 174 MB | NPU
103
+ | DeepLabV3-Plus-MobileNet | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 6.252 ms | 0 - 196 MB | NPU
104
+ | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 3.075 ms | 2 - 197 MB | NPU
105
+ | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a16 | Snapdragon® X2 Elite | 4.147 ms | 2 - 2 MB | NPU
106
+ | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a16 | Snapdragon® X Elite | 9.196 ms | 2 - 2 MB | NPU
107
+ | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a16 | Snapdragon® 8 Gen 3 Mobile | 6.12 ms | 0 - 220 MB | NPU
108
+ | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a16 | Qualcomm® QCS6490 | 33.324 ms | 1 - 4 MB | NPU
109
  | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a16 | Qualcomm® QCS8275 (Proxy) | 21.941 ms | 2 - 182 MB | NPU
110
+ | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a16 | Qualcomm® QCS8550 (Proxy) | 8.466 ms | 2 - 5 MB | NPU
111
+ | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a16 | Qualcomm® SA8775P | 8.973 ms | 2 - 184 MB | NPU
112
+ | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a16 | Qualcomm® QCS9075 | 9.799 ms | 3 - 6 MB | NPU
113
+ | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a16 | Qualcomm® QCM6690 | 111.698 ms | 2 - 232 MB | NPU
114
+ | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a16 | Qualcomm® QCS8450 (Proxy) | 12.068 ms | 2 - 221 MB | NPU
115
  | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a16 | Qualcomm® SA7255P | 21.941 ms | 2 - 182 MB | NPU
116
+ | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a16 | Qualcomm® SA8295P | 13.724 ms | 1 - 196 MB | NPU
117
+ | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a16 | Snapdragon® 8 Elite For Galaxy Mobile | 4.34 ms | 0 - 179 MB | NPU
118
+ | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a16 | Snapdragon® 7 Gen 4 Mobile | 11.905 ms | 2 - 180 MB | NPU
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
119
  | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 1.507 ms | 1 - 175 MB | NPU
120
+ | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a8 | Snapdragon® X2 Elite | 2.074 ms | 1 - 1 MB | NPU
121
+ | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a8 | Snapdragon® X Elite | 4.566 ms | 1 - 1 MB | NPU
122
+ | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a8 | Snapdragon® 8 Gen 3 Mobile | 2.898 ms | 1 - 197 MB | NPU
123
+ | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a8 | Qualcomm® QCS6490 | 15.572 ms | 1 - 3 MB | NPU
124
+ | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a8 | Qualcomm® QCS8275 (Proxy) | 10.799 ms | 1 - 169 MB | NPU
125
+ | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a8 | Qualcomm® QCS8550 (Proxy) | 4.112 ms | 1 - 96 MB | NPU
126
+ | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a8 | Qualcomm® SA8775P | 4.619 ms | 1 - 172 MB | NPU
127
+ | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a8 | Qualcomm® QCS9075 | 5.014 ms | 3 - 5 MB | NPU
128
+ | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a8 | Qualcomm® QCM6690 | 46.419 ms | 1 - 201 MB | NPU
129
+ | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a8 | Qualcomm® QCS8450 (Proxy) | 6.448 ms | 1 - 198 MB | NPU
130
+ | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a8 | Qualcomm® SA7255P | 10.799 ms | 1 - 169 MB | NPU
131
+ | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a8 | Qualcomm® SA8295P | 6.402 ms | 1 - 167 MB | NPU
132
+ | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 2.144 ms | 1 - 175 MB | NPU
133
+ | DeepLabV3-Plus-MobileNet | QNN_DLC | w8a8 | Snapdragon® 7 Gen 4 Mobile | 6.141 ms | 1 - 175 MB | NPU
134
+ | DeepLabV3-Plus-MobileNet | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 4.297 ms | 0 - 193 MB | NPU
135
+ | DeepLabV3-Plus-MobileNet | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 8.053 ms | 0 - 218 MB | NPU
136
+ | DeepLabV3-Plus-MobileNet | TFLITE | float | Qualcomm® QCS8275 (Proxy) | 58.633 ms | 0 - 170 MB | NPU
137
+ | DeepLabV3-Plus-MobileNet | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 11.441 ms | 0 - 178 MB | NPU
138
+ | DeepLabV3-Plus-MobileNet | TFLITE | float | Qualcomm® SA8775P | 17.546 ms | 0 - 174 MB | NPU
139
+ | DeepLabV3-Plus-MobileNet | TFLITE | float | Qualcomm® QCS9075 | 19.671 ms | 0 - 18 MB | NPU
140
+ | DeepLabV3-Plus-MobileNet | TFLITE | float | Qualcomm® QCS8450 (Proxy) | 18.947 ms | 0 - 217 MB | NPU
141
+ | DeepLabV3-Plus-MobileNet | TFLITE | float | Qualcomm® SA7255P | 58.633 ms | 0 - 170 MB | NPU
142
+ | DeepLabV3-Plus-MobileNet | TFLITE | float | Qualcomm® SA8295P | 19.095 ms | 0 - 175 MB | NPU
143
+ | DeepLabV3-Plus-MobileNet | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 6.252 ms | 0 - 196 MB | NPU
144
+ | DeepLabV3-Plus-MobileNet | TFLITE | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 1.526 ms | 0 - 182 MB | NPU
145
+ | DeepLabV3-Plus-MobileNet | TFLITE | w8a8 | Snapdragon® 8 Gen 3 Mobile | 2.943 ms | 0 - 205 MB | NPU
146
+ | DeepLabV3-Plus-MobileNet | TFLITE | w8a8 | Qualcomm® QCS6490 | 16.027 ms | 0 - 11 MB | NPU
147
+ | DeepLabV3-Plus-MobileNet | TFLITE | w8a8 | Qualcomm® QCS8275 (Proxy) | 10.863 ms | 0 - 173 MB | NPU
148
+ | DeepLabV3-Plus-MobileNet | TFLITE | w8a8 | Qualcomm® QCS8550 (Proxy) | 4.329 ms | 0 - 2 MB | NPU
149
+ | DeepLabV3-Plus-MobileNet | TFLITE | w8a8 | Qualcomm® SA8775P | 4.676 ms | 0 - 175 MB | NPU
150
+ | DeepLabV3-Plus-MobileNet | TFLITE | w8a8 | Qualcomm® QCS9075 | 5.175 ms | 0 - 10 MB | NPU
151
+ | DeepLabV3-Plus-MobileNet | TFLITE | w8a8 | Qualcomm® QCM6690 | 48.485 ms | 0 - 193 MB | NPU
152
+ | DeepLabV3-Plus-MobileNet | TFLITE | w8a8 | Qualcomm® QCS8450 (Proxy) | 5.644 ms | 0 - 205 MB | NPU
153
+ | DeepLabV3-Plus-MobileNet | TFLITE | w8a8 | Qualcomm® SA7255P | 10.863 ms | 0 - 173 MB | NPU
154
+ | DeepLabV3-Plus-MobileNet | TFLITE | w8a8 | Qualcomm® SA8295P | 6.259 ms | 0 - 172 MB | NPU
155
+ | DeepLabV3-Plus-MobileNet | TFLITE | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 2.096 ms | 0 - 177 MB | NPU
156
+ | DeepLabV3-Plus-MobileNet | TFLITE | w8a8 | Snapdragon® 7 Gen 4 Mobile | 6.317 ms | 0 - 178 MB | NPU
157
 
158
  ## License
159
  * The license for the original implementation of DeepLabV3-Plus-MobileNet can be found