Hugging Face's logo

qualcomm
/
DDRNet23-Slim

Image Segmentation
PyTorch
real_time
android
Model card Files
xet
 Community

DDRNet23-Slim: Optimized for Qualcomm Devices

DDRNet23Slim is a machine learning model that segments an image into semantic classes, specifically designed for road-based scenes. It is designed for the application of self-driving cars.

This is based on the implementation of DDRNet23-Slim 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 1.24.3 Download
ONNX w8a8 Universal QAIRT 2.42, ONNX Runtime 1.24.3 Download
QNN_DLC float 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 DDRNet23-Slim 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 DDRNet23-Slim on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.semantic_segmentation

Model Stats:

  • Model checkpoint: DDRNet23s_imagenet.pth
  • Inference latency: RealTime
  • Input resolution: 2048x1024
  • Number of output classes: 19
  • Number of parameters: 6.13M
  • Model size (float): 21.7 MB
  • Model size (w8a8): 6.11 MB

Performance Summary

Model Runtime Precision Chipset Inference Time (ms) Peak Memory Range (MB) Primary Compute Unit
DDRNet23-Slim ONNX float Snapdragon® 8 Elite Gen 5 Mobile 10.464 ms 21 - 246 MB NPU
DDRNet23-Slim ONNX float Snapdragon® 8 Elite Mobile 13.695 ms 8 - 206 MB NPU
DDRNet23-Slim ONNX float Snapdragon® X2 Elite 10.933 ms 22 - 22 MB NPU
DDRNet23-Slim ONNX float Snapdragon® X Elite 28.016 ms 24 - 24 MB NPU
DDRNet23-Slim ONNX float Snapdragon® X Elite 28.016 ms 24 - 24 MB NPU
DDRNet23-Slim ONNX float Snapdragon® 8 Gen 3 Mobile 19.505 ms 1 - 276 MB NPU
DDRNet23-Slim ONNX float Qualcomm® QCS8550 (Proxy) 28.367 ms 24 - 26 MB NPU
DDRNet23-Slim ONNX float Qualcomm® QCS9075 39.215 ms 24 - 51 MB NPU
DDRNet23-Slim ONNX float Snapdragon® 8 Elite For Galaxy Mobile 13.695 ms 8 - 206 MB NPU
DDRNet23-Slim ONNX w8a8 Snapdragon® 8 Elite Gen 5 Mobile 44.292 ms 80 - 272 MB NPU
DDRNet23-Slim ONNX w8a8 Snapdragon® 8 Elite Mobile 42.241 ms 86 - 280 MB NPU
DDRNet23-Slim ONNX w8a8 Snapdragon® X2 Elite 44.642 ms 109 - 109 MB NPU
DDRNet23-Slim ONNX w8a8 Snapdragon® X Elite 86.846 ms 109 - 109 MB NPU
DDRNet23-Slim ONNX w8a8 Snapdragon® X Elite 86.846 ms 109 - 109 MB NPU
DDRNet23-Slim ONNX w8a8 Snapdragon® 8 Gen 3 Mobile 44.48 ms 92 - 335 MB NPU
DDRNet23-Slim ONNX w8a8 Qualcomm® QCS6490 299.252 ms 198 - 216 MB CPU
DDRNet23-Slim ONNX w8a8 Qualcomm® QCS8550 (Proxy) 57.063 ms 86 - 316 MB NPU
DDRNet23-Slim ONNX w8a8 Qualcomm® QCS9075 65.594 ms 87 - 89 MB NPU
DDRNet23-Slim ONNX w8a8 Qualcomm® QCM6690 266.948 ms 134 - 144 MB CPU
DDRNet23-Slim ONNX w8a8 Snapdragon® 8 Elite For Galaxy Mobile 42.241 ms 86 - 280 MB NPU
DDRNet23-Slim ONNX w8a8 Snapdragon® 7 Gen 4 Mobile 251.609 ms 138 - 148 MB CPU
DDRNet23-Slim ONNX w8a8 Snapdragon® 7 Gen 4 Mobile 251.609 ms 138 - 148 MB CPU
DDRNet23-Slim QNN_DLC float Snapdragon® 8 Elite Gen 5 Mobile 10.429 ms 9 - 245 MB NPU
DDRNet23-Slim QNN_DLC float Snapdragon® 8 Elite Mobile 15.51 ms 10 - 232 MB NPU
DDRNet23-Slim QNN_DLC float Snapdragon® X2 Elite 11.803 ms 24 - 24 MB NPU
DDRNet23-Slim QNN_DLC float Snapdragon® X Elite 33.794 ms 24 - 24 MB NPU
DDRNet23-Slim QNN_DLC float Snapdragon® X Elite 33.794 ms 24 - 24 MB NPU
DDRNet23-Slim QNN_DLC float Snapdragon® 8 Gen 3 Mobile 22.187 ms 24 - 297 MB NPU
DDRNet23-Slim QNN_DLC float Qualcomm® QCS8275 (Proxy) 98.443 ms 24 - 220 MB NPU
DDRNet23-Slim QNN_DLC float Qualcomm® QCS8550 (Proxy) 32.816 ms 24 - 26 MB NPU
DDRNet23-Slim QNN_DLC float Qualcomm® SA8775P 40.454 ms 24 - 221 MB NPU
DDRNet23-Slim QNN_DLC float Qualcomm® SA8775P 40.454 ms 24 - 221 MB NPU
DDRNet23-Slim QNN_DLC float Qualcomm® SA8775P 40.454 ms 24 - 221 MB NPU
DDRNet23-Slim QNN_DLC float Qualcomm® QCS9075 54.009 ms 24 - 52 MB NPU
DDRNet23-Slim QNN_DLC float Qualcomm® QCS8450 (Proxy) 67.413 ms 5 - 282 MB NPU
DDRNet23-Slim QNN_DLC float Qualcomm® SA7255P 98.443 ms 24 - 220 MB NPU
DDRNet23-Slim QNN_DLC float Qualcomm® SA8295P 43.542 ms 24 - 230 MB NPU
DDRNet23-Slim QNN_DLC float Snapdragon® 8 Elite For Galaxy Mobile 15.51 ms 10 - 232 MB NPU
DDRNet23-Slim QNN_DLC w8a8 Snapdragon® 8 Elite Gen 5 Mobile 42.107 ms 6 - 236 MB NPU
DDRNet23-Slim QNN_DLC w8a8 Snapdragon® 8 Elite Mobile 39.967 ms 6 - 221 MB NPU
DDRNet23-Slim QNN_DLC w8a8 Snapdragon® X2 Elite 42.673 ms 6 - 6 MB NPU
DDRNet23-Slim QNN_DLC w8a8 Snapdragon® X Elite 57.964 ms 6 - 6 MB NPU
DDRNet23-Slim QNN_DLC w8a8 Snapdragon® X Elite 57.964 ms 6 - 6 MB NPU
DDRNet23-Slim QNN_DLC w8a8 Snapdragon® 8 Gen 3 Mobile 41.319 ms 6 - 248 MB NPU
DDRNet23-Slim QNN_DLC w8a8 Qualcomm® QCS6490 134.749 ms 5 - 13 MB NPU
DDRNet23-Slim QNN_DLC w8a8 Qualcomm® QCS8275 (Proxy) 106.876 ms 6 - 203 MB NPU
DDRNet23-Slim QNN_DLC w8a8 Qualcomm® QCS8550 (Proxy) 55.388 ms 6 - 146 MB NPU
DDRNet23-Slim QNN_DLC w8a8 Qualcomm® SA8775P 58.592 ms 6 - 205 MB NPU
DDRNet23-Slim QNN_DLC w8a8 Qualcomm® SA8775P 58.592 ms 6 - 205 MB NPU
DDRNet23-Slim QNN_DLC w8a8 Qualcomm® SA8775P 58.592 ms 6 - 205 MB NPU
DDRNet23-Slim QNN_DLC w8a8 Qualcomm® QCS9075 58.634 ms 6 - 14 MB NPU
DDRNet23-Slim QNN_DLC w8a8 Qualcomm® QCM6690 205.932 ms 6 - 198 MB NPU
DDRNet23-Slim QNN_DLC w8a8 Qualcomm® QCS8450 (Proxy) 60.788 ms 6 - 250 MB NPU
DDRNet23-Slim QNN_DLC w8a8 Qualcomm® SA7255P 106.876 ms 6 - 203 MB NPU
DDRNet23-Slim QNN_DLC w8a8 Qualcomm® SA8295P 63.609 ms 6 - 206 MB NPU
DDRNet23-Slim QNN_DLC w8a8 Snapdragon® 8 Elite For Galaxy Mobile 39.967 ms 6 - 221 MB NPU
DDRNet23-Slim QNN_DLC w8a8 Snapdragon® 7 Gen 4 Mobile 56.998 ms 6 - 210 MB NPU
DDRNet23-Slim QNN_DLC w8a8 Snapdragon® 7 Gen 4 Mobile 56.998 ms 6 - 210 MB NPU
DDRNet23-Slim TFLITE float Snapdragon® 8 Elite Gen 5 Mobile 10.358 ms 2 - 240 MB NPU
DDRNet23-Slim TFLITE float Snapdragon® 8 Elite Mobile 15.481 ms 1 - 225 MB NPU
DDRNet23-Slim TFLITE float Snapdragon® 8 Gen 3 Mobile 22.443 ms 2 - 285 MB NPU
DDRNet23-Slim TFLITE float Qualcomm® QCS8275 (Proxy) 98.253 ms 2 - 204 MB NPU
DDRNet23-Slim TFLITE float Qualcomm® QCS8550 (Proxy) 32.613 ms 3 - 5 MB NPU
DDRNet23-Slim TFLITE float Qualcomm® SA8775P 40.443 ms 2 - 204 MB NPU
DDRNet23-Slim TFLITE float Qualcomm® SA8775P 40.443 ms 2 - 204 MB NPU
DDRNet23-Slim TFLITE float Qualcomm® SA8775P 40.443 ms 2 - 204 MB NPU
DDRNet23-Slim TFLITE float Qualcomm® QCS9075 53.678 ms 0 - 41 MB NPU
DDRNet23-Slim TFLITE float Qualcomm® QCS8450 (Proxy) 66.866 ms 0 - 283 MB NPU
DDRNet23-Slim TFLITE float Qualcomm® SA7255P 98.253 ms 2 - 204 MB NPU
DDRNet23-Slim TFLITE float Qualcomm® SA8295P 43.494 ms 2 - 215 MB NPU
DDRNet23-Slim TFLITE float Snapdragon® 8 Elite For Galaxy Mobile 15.481 ms 1 - 225 MB NPU
DDRNet23-Slim TFLITE w8a8 Snapdragon® 8 Elite Gen 5 Mobile 21.519 ms 1 - 232 MB NPU
DDRNet23-Slim TFLITE w8a8 Snapdragon® 8 Elite Mobile 68.954 ms 0 - 215 MB NPU
DDRNet23-Slim TFLITE w8a8 Snapdragon® 8 Gen 3 Mobile 36.848 ms 1 - 249 MB NPU
DDRNet23-Slim TFLITE w8a8 Qualcomm® QCS6490 197.097 ms 9 - 77 MB NPU
DDRNet23-Slim TFLITE w8a8 Qualcomm® QCS8275 (Proxy) 95.075 ms 1 - 200 MB NPU
DDRNet23-Slim TFLITE w8a8 Qualcomm® QCS8550 (Proxy) 48.841 ms 0 - 26 MB NPU
DDRNet23-Slim TFLITE w8a8 Qualcomm® SA8775P 49.639 ms 1 - 200 MB NPU
DDRNet23-Slim TFLITE w8a8 Qualcomm® SA8775P 49.639 ms 1 - 200 MB NPU
DDRNet23-Slim TFLITE w8a8 Qualcomm® SA8775P 49.639 ms 1 - 200 MB NPU
DDRNet23-Slim TFLITE w8a8 Qualcomm® QCS9075 51.736 ms 0 - 15 MB NPU
DDRNet23-Slim TFLITE w8a8 Qualcomm® QCM6690 216.016 ms 10 - 228 MB NPU
DDRNet23-Slim TFLITE w8a8 Qualcomm® QCS8450 (Proxy) 56.707 ms 0 - 245 MB NPU
DDRNet23-Slim TFLITE w8a8 Qualcomm® SA7255P 95.075 ms 1 - 200 MB NPU
DDRNet23-Slim TFLITE w8a8 Qualcomm® SA8295P 56.07 ms 1 - 203 MB NPU
DDRNet23-Slim TFLITE w8a8 Snapdragon® 8 Elite For Galaxy Mobile 68.954 ms 0 - 215 MB NPU
DDRNet23-Slim TFLITE w8a8 Snapdragon® 7 Gen 4 Mobile 63.721 ms 9 - 208 MB NPU
DDRNet23-Slim TFLITE w8a8 Snapdragon® 7 Gen 4 Mobile 63.721 ms 9 - 208 MB NPU

License

  • The license for the original implementation of DDRNet23-Slim can be found here.

References

Community

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
Image Segmentation
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Paper for qualcomm/DDRNet23-Slim