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 2.45, ONNX Runtime 1.25.0	Download
ONNX	w8a8	Universal	QAIRT 2.45, ONNX Runtime 1.25.0	Download
QNN_DLC	float	Universal	QAIRT 2.45	Download
QNN_DLC	w8a8	Universal	QAIRT 2.45	Download
TFLITE	float	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® X2 Elite	11.706 ms	24 - 24 MB	NPU
DDRNet23-Slim	ONNX	float	Snapdragon® X Elite	27.558 ms	24 - 24 MB	NPU
DDRNet23-Slim	ONNX	float	Snapdragon® 8 Gen 3 Mobile	20.024 ms	32 - 302 MB	NPU
DDRNet23-Slim	ONNX	float	Snapdragon® 8 Gen 1 Mobile	45.839 ms	24 - 297 MB	NPU
DDRNet23-Slim	ONNX	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	27.955 ms	0 - 212 MB	NPU
DDRNet23-Slim	ONNX	float	Qualcomm® QCS8450	45.839 ms	24 - 297 MB	NPU
DDRNet23-Slim	ONNX	float	Qualcomm® Dragonwing™ IQ-9075	41.647 ms	24 - 51 MB	NPU
DDRNet23-Slim	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	11.414 ms	5 - 219 MB	NPU
DDRNet23-Slim	ONNX	float	Snapdragon® 8 Elite Mobile	14.977 ms	7 - 194 MB	NPU
DDRNet23-Slim	ONNX	float	Qualcomm® Dragonwing™ Q-8750	14.977 ms	7 - 194 MB	NPU
DDRNet23-Slim	ONNX	float	Qualcomm® Dragonwing™ IQ-X7181	27.558 ms	24 - 24 MB	NPU
DDRNet23-Slim	ONNX	w8a8	Snapdragon® X2 Elite	43.774 ms	7 - 7 MB	NPU
DDRNet23-Slim	ONNX	w8a8	Snapdragon® X Elite	55.955 ms	7 - 7 MB	NPU
DDRNet23-Slim	ONNX	w8a8	Snapdragon® 8 Gen 3 Mobile	39.605 ms	7 - 253 MB	NPU
DDRNet23-Slim	ONNX	w8a8	Snapdragon® 8 Gen 1 Mobile	55.21 ms	8 - 255 MB	NPU
DDRNet23-Slim	ONNX	w8a8	Qualcomm® Dragonwing™ QCS6490	149.785 ms	6 - 9 MB	NPU
DDRNet23-Slim	ONNX	w8a8	Qualcomm® Dragonwing™ QCS8550 (Proxy)	53.345 ms	6 - 9 MB	NPU
DDRNet23-Slim	ONNX	w8a8	Qualcomm® QCS8450	55.21 ms	8 - 255 MB	NPU
DDRNet23-Slim	ONNX	w8a8	Snapdragon® 8 Elite Mobile	40.987 ms	1 - 199 MB	NPU
DDRNet23-Slim	ONNX	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	43.375 ms	2 - 204 MB	NPU
DDRNet23-Slim	ONNX	w8a8	Qualcomm® Dragonwing™ IQ-9075	58.838 ms	6 - 9 MB	NPU
DDRNet23-Slim	ONNX	w8a8	Snapdragon® 7 Gen 4 Mobile	53.663 ms	7 - 208 MB	NPU
DDRNet23-Slim	ONNX	w8a8	Qualcomm® Dragonwing™ Q-6690	196.435 ms	6 - 206 MB	NPU
DDRNet23-Slim	ONNX	w8a8	Qualcomm® Dragonwing™ Q-7790	53.663 ms	7 - 208 MB	NPU
DDRNet23-Slim	ONNX	w8a8	Qualcomm® Dragonwing™ Q-8750	40.987 ms	1 - 199 MB	NPU
DDRNet23-Slim	ONNX	w8a8	Qualcomm® Dragonwing™ IQ-X7181	55.955 ms	7 - 7 MB	NPU
DDRNet23-Slim	QNN_DLC	float	Snapdragon® X2 Elite	11.75 ms	24 - 24 MB	NPU
DDRNet23-Slim	QNN_DLC	float	Snapdragon® X Elite	33.869 ms	24 - 24 MB	NPU
DDRNet23-Slim	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	22.398 ms	24 - 297 MB	NPU
DDRNet23-Slim	QNN_DLC	float	Snapdragon® 8 Gen 1 Mobile	66.647 ms	0 - 278 MB	NPU
DDRNet23-Slim	QNN_DLC	float	Qualcomm® QCS8275	98.47 ms	24 - 221 MB	NPU
DDRNet23-Slim	QNN_DLC	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	32.9 ms	24 - 173 MB	NPU
DDRNet23-Slim	QNN_DLC	float	Qualcomm® SA8775P	40.434 ms	24 - 221 MB	NPU
DDRNet23-Slim	QNN_DLC	float	Qualcomm® SA8650P	40.434 ms	24 - 221 MB	NPU
DDRNet23-Slim	QNN_DLC	float	Qualcomm® SA8255P	40.434 ms	24 - 221 MB	NPU
DDRNet23-Slim	QNN_DLC	float	Qualcomm® QCS8450	66.647 ms	0 - 278 MB	NPU
DDRNet23-Slim	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-9075	41.719 ms	24 - 52 MB	NPU
DDRNet23-Slim	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	10.517 ms	20 - 259 MB	NPU
DDRNet23-Slim	QNN_DLC	float	Qualcomm® SA7255P	98.47 ms	24 - 221 MB	NPU
DDRNet23-Slim	QNN_DLC	float	Snapdragon® 8 Elite Mobile	15.457 ms	15 - 235 MB	NPU
DDRNet23-Slim	QNN_DLC	float	Qualcomm® SA8295P	43.495 ms	24 - 229 MB	NPU
DDRNet23-Slim	QNN_DLC	float	Qualcomm® Dragonwing™ Q-8750	15.457 ms	15 - 235 MB	NPU
DDRNet23-Slim	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-X7181	33.869 ms	24 - 24 MB	NPU
DDRNet23-Slim	QNN_DLC	w8a8	Snapdragon® X2 Elite	29.317 ms	6 - 6 MB	NPU
DDRNet23-Slim	QNN_DLC	w8a8	Snapdragon® X Elite	58.853 ms	6 - 6 MB	NPU
DDRNet23-Slim	QNN_DLC	w8a8	Snapdragon® 8 Gen 3 Mobile	41.863 ms	6 - 251 MB	NPU
DDRNet23-Slim	QNN_DLC	w8a8	Snapdragon® 8 Gen 1 Mobile	80.52 ms	7 - 254 MB	NPU
DDRNet23-Slim	QNN_DLC	w8a8	Qualcomm® QCS8275	107.872 ms	6 - 205 MB	NPU
DDRNet23-Slim	QNN_DLC	w8a8	Qualcomm® Dragonwing™ QCS8550 (Proxy)	56.214 ms	6 - 8 MB	NPU
DDRNet23-Slim	QNN_DLC	w8a8	Qualcomm® SA8775P	56.908 ms	6 - 206 MB	NPU
DDRNet23-Slim	QNN_DLC	w8a8	Qualcomm® SA8650P	56.908 ms	6 - 206 MB	NPU
DDRNet23-Slim	QNN_DLC	w8a8	Qualcomm® SA8255P	56.908 ms	6 - 206 MB	NPU
DDRNet23-Slim	QNN_DLC	w8a8	Qualcomm® QCS8450	80.52 ms	7 - 254 MB	NPU
DDRNet23-Slim	QNN_DLC	w8a8	Snapdragon® 8 Elite Mobile	39.65 ms	0 - 218 MB	NPU
DDRNet23-Slim	QNN_DLC	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	28.685 ms	6 - 240 MB	NPU
DDRNet23-Slim	QNN_DLC	w8a8	Qualcomm® Dragonwing™ IQ-9075	57.505 ms	6 - 14 MB	NPU
DDRNet23-Slim	QNN_DLC	w8a8	Qualcomm® SA7255P	107.872 ms	6 - 205 MB	NPU
DDRNet23-Slim	QNN_DLC	w8a8	Qualcomm® SA8295P	66.582 ms	6 - 208 MB	NPU
DDRNet23-Slim	QNN_DLC	w8a8	Qualcomm® Dragonwing™ Q-8750	39.65 ms	0 - 218 MB	NPU
DDRNet23-Slim	QNN_DLC	w8a8	Qualcomm® Dragonwing™ IQ-X7181	58.853 ms	6 - 6 MB	NPU
DDRNet23-Slim	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	22.433 ms	2 - 286 MB	NPU
DDRNet23-Slim	TFLITE	float	Snapdragon® 8 Gen 1 Mobile	66.498 ms	3 - 287 MB	NPU
DDRNet23-Slim	TFLITE	float	Qualcomm® QCS8275	98.332 ms	3 - 204 MB	NPU
DDRNet23-Slim	TFLITE	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	33.091 ms	2 - 5 MB	NPU
DDRNet23-Slim	TFLITE	float	Qualcomm® SA8775P	40.417 ms	3 - 206 MB	NPU
DDRNet23-Slim	TFLITE	float	Qualcomm® SA8650P	40.417 ms	3 - 206 MB	NPU
DDRNet23-Slim	TFLITE	float	Qualcomm® SA8255P	40.417 ms	3 - 206 MB	NPU
DDRNet23-Slim	TFLITE	float	Qualcomm® QCS8450	66.498 ms	3 - 287 MB	NPU
DDRNet23-Slim	TFLITE	float	Qualcomm® Dragonwing™ IQ-9075	41.467 ms	0 - 41 MB	NPU
DDRNet23-Slim	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	10.448 ms	2 - 242 MB	NPU
DDRNet23-Slim	TFLITE	float	Qualcomm® SA7255P	98.332 ms	3 - 204 MB	NPU
DDRNet23-Slim	TFLITE	float	Snapdragon® 8 Elite Mobile	15.441 ms	2 - 225 MB	NPU
DDRNet23-Slim	TFLITE	float	Qualcomm® SA8295P	43.444 ms	3 - 215 MB	NPU
DDRNet23-Slim	TFLITE	float	Qualcomm® Dragonwing™ Q-8750	15.441 ms	2 - 225 MB	NPU

License

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

References

Community

Join our AI Hub Slack community to collaborate, post questions and learn more about on-device AI.
For questions or feedback please reach out to us.

Downloads last month: -; Downloads are not tracked for this model. How to track

Paper for qualcomm/DDRNet23-Slim

Deep Dual-resolution Networks for Real-time and Accurate Semantic Segmentation of Road Scenes

Paper • 2101.06085 • Published Jan 15, 2021