Hugging Face's logo

qualcomm
/
DeepLabV3-Plus-MobileNet

Image Segmentation
PyTorch
android
Model card Files
xet
 Community
DeepLabV3-Plus-MobileNet / README.md
qaihm-bot's picture
qaihm-bot
v0.55.0
6b5f9e8 verified
preview code
|
raw
history blame contribute delete
16.7 kB
metadata 
library_name: pytorch
license: other
tags:
  - android
pipeline_tag: image-segmentation

DeepLabV3-Plus-MobileNet: Optimized for Qualcomm Devices

DeepLabV3 is designed for semantic segmentation at multiple scales, trained on the various datasets. It uses MobileNet as a backbone.

This is based on the implementation of DeepLabV3-Plus-MobileNet found here. This repository contains pre-exported model files optimized for Qualcomm® devices. You can use the Qualcomm® AI Hub Models library to export with custom configurations. More details on model performance across various devices, can be found here.

Qualcomm AI Hub Models uses Qualcomm AI Hub Workbench to compile, profile, and evaluate this model. Sign up to run these models on a hosted Qualcomm® device.

Getting Started

There are two ways to deploy this model on your device:

Option 1: Download Pre-Exported Models

Below are pre-exported model assets ready for deployment.

Runtime Precision Chipset SDK Versions Download
ONNX float Universal QAIRT 2.42, ONNX Runtime 1.25.0 Download
ONNX w8a16 Universal QAIRT 2.42, ONNX Runtime 1.25.0 Download
ONNX w8a8 Universal QAIRT 2.42, ONNX Runtime 1.25.0 Download
QNN_DLC float Universal QAIRT 2.45 Download
QNN_DLC w8a16 Universal QAIRT 2.45 Download
QNN_DLC w8a8 Universal QAIRT 2.45 Download
TFLITE float Universal QAIRT 2.45 Download
TFLITE w8a8 Universal QAIRT 2.45 Download

For more device-specific assets and performance metrics, visit DeepLabV3-Plus-MobileNet on Qualcomm® AI Hub.

Option 2: Export with Custom Configurations

Use the Qualcomm® AI Hub Models Python library to compile and export the model with your own:

  • Custom weights (e.g., fine-tuned checkpoints)
  • Custom input shapes
  • Target device and runtime configurations

This option is ideal if you need to customize the model beyond the default configuration provided here.

See our repository for DeepLabV3-Plus-MobileNet on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.semantic_segmentation

Model Stats:

  • Model checkpoint: VOC2012
  • Input resolution: 513x513
  • Number of output classes: 21
  • Number of parameters: 5.80M
  • Model size (float): 22.2 MB
  • Model size (w8a16): 6.67 MB

Performance Summary

Model Runtime Precision Chipset Inference Time (ms) Peak Memory Range (MB) Primary Compute Unit
DeepLabV3-Plus-MobileNet ONNX float Snapdragon® 8 Elite Gen 5 Mobile 4.301 ms 1 - 176 MB NPU
DeepLabV3-Plus-MobileNet ONNX float Snapdragon® X2 Elite 5.389 ms 209 - 209 MB NPU
DeepLabV3-Plus-MobileNet ONNX float Snapdragon® X Elite 10.906 ms 146 - 146 MB NPU
DeepLabV3-Plus-MobileNet ONNX float Snapdragon® 8 Gen 3 Mobile 7.197 ms 4 - 214 MB NPU
DeepLabV3-Plus-MobileNet ONNX float Qualcomm® QCS8550 (Proxy) 10.352 ms 0 - 44 MB NPU
DeepLabV3-Plus-MobileNet ONNX float Snapdragon® 8 Elite For Galaxy Mobile 5.893 ms 1 - 170 MB NPU
DeepLabV3-Plus-MobileNet ONNX float Qualcomm® QCS9075 18.037 ms 3 - 48 MB NPU
DeepLabV3-Plus-MobileNet ONNX float Qualcomm® QCS8750 5.893 ms 1 - 170 MB NPU
DeepLabV3-Plus-MobileNet ONNX float Qualcomm® QCS7181 10.906 ms 146 - 146 MB NPU
DeepLabV3-Plus-MobileNet ONNX w8a16 Snapdragon® 8 Elite Gen 5 Mobile 2.948 ms 0 - 213 MB NPU
DeepLabV3-Plus-MobileNet ONNX w8a16 Snapdragon® X2 Elite 3.689 ms 211 - 211 MB NPU
DeepLabV3-Plus-MobileNet ONNX w8a16 Snapdragon® X Elite 7.418 ms 180 - 180 MB NPU
DeepLabV3-Plus-MobileNet ONNX w8a16 Snapdragon® 8 Gen 3 Mobile 5.074 ms 2 - 223 MB NPU
DeepLabV3-Plus-MobileNet ONNX w8a16 Qualcomm® QCS6490 1218.715 ms 89 - 93 MB CPU
DeepLabV3-Plus-MobileNet ONNX w8a16 Qualcomm® QCS8550 (Proxy) 7.043 ms 0 - 40 MB NPU
DeepLabV3-Plus-MobileNet ONNX w8a16 Snapdragon® 7 Gen 4 Mobile 577.148 ms 92 - 100 MB CPU
DeepLabV3-Plus-MobileNet ONNX w8a16 Snapdragon® 8 Elite For Galaxy Mobile 3.913 ms 0 - 186 MB NPU
DeepLabV3-Plus-MobileNet ONNX w8a16 Qualcomm® QCM6690 593.153 ms 87 - 95 MB CPU
DeepLabV3-Plus-MobileNet ONNX w8a16 Qualcomm® QCS9075 8.512 ms 0 - 50 MB NPU
DeepLabV3-Plus-MobileNet ONNX w8a16 Qualcomm® QCS7790 577.148 ms 92 - 100 MB CPU
DeepLabV3-Plus-MobileNet ONNX w8a16 Qualcomm® QCS8750 3.913 ms 0 - 186 MB NPU
DeepLabV3-Plus-MobileNet ONNX w8a16 Qualcomm® QCS7181 7.418 ms 180 - 180 MB NPU
DeepLabV3-Plus-MobileNet ONNX w8a8 Snapdragon® 8 Elite Gen 5 Mobile 1.458 ms 0 - 181 MB NPU
DeepLabV3-Plus-MobileNet ONNX w8a8 Snapdragon® X2 Elite 1.659 ms 211 - 211 MB NPU
DeepLabV3-Plus-MobileNet ONNX w8a8 Snapdragon® X Elite 3.673 ms 181 - 181 MB NPU
DeepLabV3-Plus-MobileNet ONNX w8a8 Snapdragon® 8 Gen 3 Mobile 2.375 ms 0 - 203 MB NPU
DeepLabV3-Plus-MobileNet ONNX w8a8 Qualcomm® QCS8550 (Proxy) 3.541 ms 0 - 124 MB NPU
DeepLabV3-Plus-MobileNet ONNX w8a8 Qualcomm® QCS9075 4.276 ms 0 - 51 MB NPU
DeepLabV3-Plus-MobileNet ONNX w8a8 Snapdragon® 8 Elite For Galaxy Mobile 1.888 ms 0 - 180 MB NPU
DeepLabV3-Plus-MobileNet ONNX w8a8 Qualcomm® QCS8750 1.888 ms 0 - 180 MB NPU
DeepLabV3-Plus-MobileNet ONNX w8a8 Qualcomm® QCS7181 3.673 ms 181 - 181 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC float Snapdragon® 8 Elite Gen 5 Mobile 4.306 ms 3 - 184 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC float Snapdragon® X2 Elite 5.575 ms 3 - 3 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC float Snapdragon® X Elite 12.367 ms 3 - 3 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC float Snapdragon® 8 Gen 3 Mobile 8.084 ms 3 - 211 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC float Qualcomm® QCS8275 58.614 ms 2 - 167 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC float Qualcomm® QCS8550 (Proxy) 11.664 ms 3 - 5 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC float Qualcomm® SA8775P 17.503 ms 1 - 169 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC float Qualcomm® SA8650P 17.503 ms 1 - 169 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC float Qualcomm® SA8255P 17.503 ms 1 - 169 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC float Snapdragon® 8 Elite For Galaxy Mobile 6.287 ms 0 - 175 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC float Qualcomm® QCS8450 (Proxy) 19.821 ms 3 - 209 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC float Qualcomm® SA7255P 58.614 ms 2 - 167 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC float Qualcomm® QCS9075 20.234 ms 3 - 8 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC float Qualcomm® SA8295P 19.572 ms 3 - 172 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC float Qualcomm® QCS8750 6.287 ms 0 - 175 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC float Qualcomm® QCS7181 12.367 ms 3 - 3 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC w8a16 Snapdragon® 8 Elite Gen 5 Mobile 3.073 ms 2 - 197 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC w8a16 Snapdragon® X2 Elite 4.318 ms 2 - 2 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC w8a16 Snapdragon® X Elite 9.147 ms 2 - 2 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC w8a16 Snapdragon® 8 Gen 3 Mobile 6.125 ms 0 - 213 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC w8a16 Qualcomm® QCS6490 33.602 ms 1 - 4 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC w8a16 Qualcomm® QCS8275 21.977 ms 2 - 182 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC w8a16 Qualcomm® QCS8550 (Proxy) 8.437 ms 2 - 4 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC w8a16 Qualcomm® SA8775P 8.923 ms 2 - 183 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC w8a16 Qualcomm® SA8650P 8.923 ms 2 - 183 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC w8a16 Qualcomm® SA8255P 8.923 ms 2 - 183 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC w8a16 Snapdragon® 7 Gen 4 Mobile 11.987 ms 2 - 180 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC w8a16 Snapdragon® 8 Elite For Galaxy Mobile 4.374 ms 2 - 181 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC w8a16 Qualcomm® QCM6690 106.645 ms 2 - 234 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC w8a16 Qualcomm® SA7255P 21.977 ms 2 - 182 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC w8a16 Qualcomm® SA8295P 13.706 ms 2 - 197 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC w8a16 Qualcomm® QCS9075 9.713 ms 1 - 4 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC w8a16 Qualcomm® QCS8450 (Proxy) 12.085 ms 2 - 214 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC w8a16 Qualcomm® QCS7790 11.987 ms 2 - 180 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC w8a16 Qualcomm® QCS8750 4.374 ms 2 - 181 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC w8a16 Qualcomm® QCS7181 9.147 ms 2 - 2 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC w8a8 Snapdragon® 8 Elite Gen 5 Mobile 1.615 ms 0 - 175 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC w8a8 Snapdragon® X2 Elite 2.11 ms 1 - 1 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC w8a8 Snapdragon® X Elite 4.668 ms 1 - 1 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC w8a8 Snapdragon® 8 Gen 3 Mobile 2.905 ms 0 - 193 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC w8a8 Qualcomm® QCS6490 15.867 ms 1 - 3 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC w8a8 Qualcomm® QCS8275 10.79 ms 1 - 169 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC w8a8 Qualcomm® QCS8550 (Proxy) 4.173 ms 1 - 3 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC w8a8 Qualcomm® SA8775P 4.713 ms 1 - 171 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC w8a8 Qualcomm® SA8650P 4.713 ms 1 - 171 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC w8a8 Qualcomm® SA8255P 4.713 ms 1 - 171 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC w8a8 Qualcomm® QCS9075 5.123 ms 0 - 3 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC w8a8 Qualcomm® QCS8450 (Proxy) 6.638 ms 1 - 196 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC w8a8 Snapdragon® 7 Gen 4 Mobile 6.286 ms 1 - 176 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC w8a8 Snapdragon® 8 Elite For Galaxy Mobile 2.184 ms 1 - 173 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC w8a8 Qualcomm® QCM6690 49.045 ms 1 - 200 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC w8a8 Qualcomm® SA8295P 6.411 ms 1 - 168 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC w8a8 Qualcomm® SA7255P 10.79 ms 1 - 169 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC w8a8 Qualcomm® QCS7790 6.286 ms 1 - 176 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC w8a8 Qualcomm® QCS8750 2.184 ms 1 - 173 MB NPU
DeepLabV3-Plus-MobileNet QNN_DLC w8a8 Qualcomm® QCS7181 4.668 ms 1 - 1 MB NPU
DeepLabV3-Plus-MobileNet TFLITE float Snapdragon® 8 Elite Gen 5 Mobile 4.314 ms 0 - 183 MB NPU
DeepLabV3-Plus-MobileNet TFLITE float Snapdragon® 8 Gen 3 Mobile 8.089 ms 0 - 217 MB NPU
DeepLabV3-Plus-MobileNet TFLITE float Qualcomm® QCS8275 58.657 ms 0 - 169 MB NPU
DeepLabV3-Plus-MobileNet TFLITE float Qualcomm® QCS8550 (Proxy) 11.51 ms 0 - 3 MB NPU
DeepLabV3-Plus-MobileNet TFLITE float Qualcomm® SA8775P 17.574 ms 0 - 171 MB NPU
DeepLabV3-Plus-MobileNet TFLITE float Qualcomm® SA8650P 17.574 ms 0 - 171 MB NPU
DeepLabV3-Plus-MobileNet TFLITE float Qualcomm® SA8255P 17.574 ms 0 - 171 MB NPU
DeepLabV3-Plus-MobileNet TFLITE float Snapdragon® 8 Elite For Galaxy Mobile 6.29 ms 0 - 176 MB NPU
DeepLabV3-Plus-MobileNet TFLITE float Qualcomm® QCS8450 (Proxy) 19.81 ms 1 - 210 MB NPU
DeepLabV3-Plus-MobileNet TFLITE float Qualcomm® SA7255P 58.657 ms 0 - 169 MB NPU
DeepLabV3-Plus-MobileNet TFLITE float Qualcomm® QCS9075 19.569 ms 0 - 18 MB NPU
DeepLabV3-Plus-MobileNet TFLITE float Qualcomm® SA8295P 19.599 ms 0 - 169 MB NPU
DeepLabV3-Plus-MobileNet TFLITE float Qualcomm® QCS8750 6.29 ms 0 - 176 MB NPU
DeepLabV3-Plus-MobileNet TFLITE w8a8 Snapdragon® 8 Elite Gen 5 Mobile 1.584 ms 0 - 183 MB NPU
DeepLabV3-Plus-MobileNet TFLITE w8a8 Snapdragon® 8 Gen 3 Mobile 2.939 ms 0 - 204 MB NPU
DeepLabV3-Plus-MobileNet TFLITE w8a8 Qualcomm® QCS6490 15.848 ms 0 - 11 MB NPU
DeepLabV3-Plus-MobileNet TFLITE w8a8 Qualcomm® QCS8275 10.806 ms 0 - 174 MB NPU
DeepLabV3-Plus-MobileNet TFLITE w8a8 Qualcomm® QCS8550 (Proxy) 4.366 ms 0 - 2 MB NPU
DeepLabV3-Plus-MobileNet TFLITE w8a8 Qualcomm® SA8775P 4.611 ms 0 - 176 MB NPU
DeepLabV3-Plus-MobileNet TFLITE w8a8 Qualcomm® SA8650P 4.611 ms 0 - 176 MB NPU
DeepLabV3-Plus-MobileNet TFLITE w8a8 Qualcomm® SA8255P 4.611 ms 0 - 176 MB NPU
DeepLabV3-Plus-MobileNet TFLITE w8a8 Qualcomm® QCS9075 5.247 ms 0 - 10 MB NPU
DeepLabV3-Plus-MobileNet TFLITE w8a8 Qualcomm® QCS8450 (Proxy) 5.702 ms 0 - 200 MB NPU
DeepLabV3-Plus-MobileNet TFLITE w8a8 Snapdragon® 7 Gen 4 Mobile 6.361 ms 0 - 177 MB NPU
DeepLabV3-Plus-MobileNet TFLITE w8a8 Snapdragon® 8 Elite For Galaxy Mobile 2.084 ms 0 - 174 MB NPU
DeepLabV3-Plus-MobileNet TFLITE w8a8 Qualcomm® QCM6690 54.262 ms 0 - 196 MB NPU
DeepLabV3-Plus-MobileNet TFLITE w8a8 Qualcomm® SA8295P 6.242 ms 0 - 172 MB NPU
DeepLabV3-Plus-MobileNet TFLITE w8a8 Qualcomm® SA7255P 10.806 ms 0 - 174 MB NPU
DeepLabV3-Plus-MobileNet TFLITE w8a8 Qualcomm® QCS7790 6.361 ms 0 - 177 MB NPU
DeepLabV3-Plus-MobileNet TFLITE w8a8 Qualcomm® QCS8750 2.084 ms 0 - 174 MB NPU

License

  • The license for the original implementation of DeepLabV3-Plus-MobileNet can be found here.

References

Community