Hugging Peers

DDRNet23-Slim

☆ Save
by qualcomm image-segmentation pytorch mit
81 downloads 0 likes on Hugging Face updated

Model card

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 2.50, ONNX Runtime 1.30.0 Download
ONNX w8a8 Universal QAIRT 2.50, ONNX Runtime 1.30.0 Download
QNN_DLC float Universal QAIRT 2.50 Download
QNN_DLC w8a8 Universal QAIRT 2.50 Download
TFLITE float Universal QAIRT 2.50 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:

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

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 For Galaxy Mobile 11.104 ms 5 - 206 MB NPU
DDRNet23-Slim ONNX float Snapdragon® 8 Elite For Galaxy Mobile 14.712 ms 6 - 184 MB NPU
DDRNet23-Slim ONNX float Snapdragon® X2 Elite 11.354 ms 24 - 24 MB NPU
DDRNet23-Slim ONNX float Snapdragon® X Elite 26.314 ms 24 - 24 MB NPU
DDRNet23-Slim ONNX float Snapdragon® 8 Gen 3 Mobile 19.643 ms 29 - 289 MB NPU
DDRNet23-Slim ONNX float Snapdragon® 8 Gen 1 Mobile 38.415 ms 29 - 291 MB NPU
DDRNet23-Slim ONNX float Qualcomm® Dragonwing™ IQ-8275 33.969 ms 24 - 52 MB NPU
DDRNet23-Slim ONNX float Qualcomm® Dragonwing™ QCS8550 (Proxy) 27.402 ms 24 - 40 MB NPU
DDRNet23-Slim ONNX float Qualcomm® QCS8450 38.415 ms 29 - 291 MB NPU
DDRNet23-Slim ONNX float Qualcomm® Dragonwing™ IQ-9075 35.471 ms 24 - 51 MB NPU
DDRNet23-Slim ONNX float Qualcomm® Dragonwing™ IQ-X7181 26.314 ms 24 - 24 MB NPU
DDRNet23-Slim ONNX float Qualcomm® Dragonwing™ Q-8750 14.712 ms 6 - 184 MB NPU
DDRNet23-Slim ONNX w8a8 Snapdragon® 8 Elite Gen 5 For Galaxy Mobile 26.834 ms 2 - 199 MB NPU
DDRNet23-Slim ONNX w8a8 Snapdragon® 8 Elite For Galaxy Mobile 37.433 ms 1 - 190 MB NPU
DDRNet23-Slim ONNX w8a8 Snapdragon® X2 Elite 27.105 ms 7 - 7 MB NPU
DDRNet23-Slim ONNX w8a8 Snapdragon® X Elite 53.94 ms 7 - 7 MB NPU
DDRNet23-Slim ONNX w8a8 Snapdragon® 8 Gen 3 Mobile 37.856 ms 2 - 240 MB NPU
DDRNet23-Slim ONNX w8a8 Snapdragon® 8 Gen 1 Mobile 53.595 ms 8 - 247 MB NPU
DDRNet23-Slim ONNX w8a8 Qualcomm® Dragonwing™ IQ-8275 49.623 ms 6 - 10 MB NPU
DDRNet23-Slim ONNX w8a8 Qualcomm® Dragonwing™ QCS8550 (Proxy) 51.239 ms 6 - 83 MB NPU
DDRNet23-Slim ONNX w8a8 Qualcomm® QCS8450 53.595 ms 8 - 247 MB NPU
DDRNet23-Slim ONNX w8a8 Qualcomm® Dragonwing™ IQ-9075 52.722 ms 6 - 9 MB NPU
DDRNet23-Slim ONNX w8a8 Qualcomm® Dragonwing™ IQ-X7181 53.94 ms 7 - 7 MB NPU
DDRNet23-Slim ONNX w8a8 Qualcomm® Dragonwing™ Q-8750 37.433 ms 1 - 190 MB NPU
DDRNet23-Slim QNN_DLC float Snapdragon® 8 Elite Gen 5 For Galaxy Mobile 10.989 ms 22 - 243 MB NPU
DDRNet23-Slim QNN_DLC float Snapdragon® 8 Elite For Galaxy Mobile 15.862 ms 12 - 217 MB NPU
DDRNet23-Slim QNN_DLC float Snapdragon® X2 Elite 14.66 ms 24 - 24 MB NPU
DDRNet23-Slim QNN_DLC float Snapdragon® X Elite 33.855 ms 24 - 24 MB NPU
DDRNet23-Slim QNN_DLC float Snapdragon® 8 Gen 3 Mobile 22.473 ms 24 - 290 MB NPU
DDRNet23-Slim QNN_DLC float Snapdragon® 8 Gen 1 Mobile 42.056 ms 5 - 272 MB NPU
DDRNet23-Slim QNN_DLC float Qualcomm® Dragonwing™ IQ-8275 38.175 ms 24 - 53 MB NPU
DDRNet23-Slim QNN_DLC float Qualcomm® Dragonwing™ QCS8550 (Proxy) 32.777 ms 24 - 26 MB NPU
DDRNet23-Slim QNN_DLC float Qualcomm® SA8775P 40.287 ms 24 - 213 MB NPU
DDRNet23-Slim QNN_DLC float Qualcomm® SA8650P 40.287 ms 24 - 213 MB NPU
DDRNet23-Slim QNN_DLC float Qualcomm® SA8255P 40.287 ms 24 - 213 MB NPU
DDRNet23-Slim QNN_DLC float Qualcomm® QCS8450 42.056 ms 5 - 272 MB NPU
DDRNet23-Slim QNN_DLC float Qualcomm® Dragonwing™ IQ-9075 41.186 ms 26 - 54 MB NPU
DDRNet23-Slim QNN_DLC float Qualcomm® Dragonwing™ IQ-X7181 33.855 ms 24 - 24 MB NPU
DDRNet23-Slim QNN_DLC float Qualcomm® Dragonwing™ Q-8750 15.862 ms 12 - 217 MB NPU
DDRNet23-Slim QNN_DLC float Qualcomm® SA8295P 43.428 ms 24 - 221 MB NPU
DDRNet23-Slim QNN_DLC w8a8 Snapdragon® 8 Elite Gen 5 For Galaxy Mobile 28.652 ms 6 - 227 MB NPU
DDRNet23-Slim QNN_DLC w8a8 Snapdragon® 8 Elite For Galaxy Mobile 39.752 ms 6 - 211 MB NPU
DDRNet23-Slim QNN_DLC w8a8 Snapdragon® X2 Elite 29.262 ms 6 - 6 MB NPU
DDRNet23-Slim QNN_DLC w8a8 Snapdragon® X Elite 59.186 ms 6 - 6 MB NPU
DDRNet23-Slim QNN_DLC w8a8 Snapdragon® 8 Gen 3 Mobile 42.321 ms 6 - 247 MB NPU
DDRNet23-Slim QNN_DLC w8a8 Snapdragon® 8 Gen 1 Mobile 56.867 ms 6 - 245 MB NPU
DDRNet23-Slim QNN_DLC w8a8 Qualcomm® Dragonwing™ IQ-8275 55.131 ms 6 - 16 MB NPU
DDRNet23-Slim QNN_DLC w8a8 Qualcomm® Dragonwing™ QCS8550 (Proxy) 56.133 ms 6 - 8 MB NPU
DDRNet23-Slim QNN_DLC w8a8 Qualcomm® QCS8450 56.867 ms 6 - 245 MB NPU
DDRNet23-Slim QNN_DLC w8a8 Qualcomm® Dragonwing™ IQ-9075 57.461 ms 6 - 15 MB NPU
DDRNet23-Slim QNN_DLC w8a8 Qualcomm® Dragonwing™ IQ-X7181 59.186 ms 6 - 6 MB NPU
DDRNet23-Slim QNN_DLC w8a8 Qualcomm® Dragonwing™ Q-8750 39.752 ms 6 - 211 MB NPU
DDRNet23-Slim QNN_DLC w8a8 Qualcomm® SA8295P 64.131 ms 6 - 200 MB NPU
DDRNet23-Slim TFLITE float Snapdragon® 8 Elite Gen 5 For Galaxy Mobile 11.032 ms 2 - 225 MB NPU
DDRNet23-Slim TFLITE float Snapdragon® 8 Elite For Galaxy Mobile 15.882 ms 1 - 212 MB NPU
DDRNet23-Slim TFLITE float Snapdragon® 8 Gen 3 Mobile 22.584 ms 10 - 282 MB NPU
DDRNet23-Slim TFLITE float Snapdragon® 8 Gen 1 Mobile 42.392 ms 3 - 278 MB NPU
DDRNet23-Slim TFLITE float Qualcomm® Dragonwing™ IQ-8275 38.644 ms 0 - 42 MB NPU
DDRNet23-Slim TFLITE float Qualcomm® Dragonwing™ QCS8550 (Proxy) 32.658 ms 2 - 5 MB NPU
DDRNet23-Slim TFLITE float Qualcomm® QCS8450 42.392 ms 3 - 278 MB NPU
DDRNet23-Slim TFLITE float Qualcomm® Dragonwing™ IQ-9075 41.083 ms 0 - 41 MB NPU
DDRNet23-Slim TFLITE float Qualcomm® Dragonwing™ Q-8750 15.882 ms 1 - 212 MB NPU
DDRNet23-Slim TFLITE float Qualcomm® SA8295P 43.174 ms 2 - 203 MB NPU

License

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

References

Community