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 1.24.1	Download
ONNX	w8a8	Universal	QAIRT 2.42, ONNX Runtime 1.24.1	Download
QNN_DLC	float	Universal	QAIRT 2.43	Download
QNN_DLC	w8a8	Universal	QAIRT 2.43	Download
TFLITE	float	Universal	QAIRT 2.43, TFLite 2.17.0	Download
TFLITE	w8a8	Universal	QAIRT 2.43, TFLite 2.17.0	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	31 - 261 MB	NPU
DDRNet23-Slim	ONNX	float	Snapdragon® X2 Elite	10.899 ms	22 - 22 MB	NPU
DDRNet23-Slim	ONNX	float	Snapdragon® X Elite	28.085 ms	24 - 24 MB	NPU
DDRNet23-Slim	ONNX	float	Snapdragon® 8 Gen 3 Mobile	19.794 ms	0 - 277 MB	NPU
DDRNet23-Slim	ONNX	float	Qualcomm® QCS8550 (Proxy)	29.255 ms	24 - 27 MB	NPU
DDRNet23-Slim	ONNX	float	Qualcomm® QCS9075	39.446 ms	24 - 51 MB	NPU
DDRNet23-Slim	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	13.531 ms	7 - 201 MB	NPU
DDRNet23-Slim	ONNX	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	44.105 ms	58 - 250 MB	NPU
DDRNet23-Slim	ONNX	w8a8	Snapdragon® X2 Elite	45.035 ms	109 - 109 MB	NPU
DDRNet23-Slim	ONNX	w8a8	Snapdragon® X Elite	64.712 ms	109 - 109 MB	NPU
DDRNet23-Slim	ONNX	w8a8	Snapdragon® 8 Gen 3 Mobile	43.106 ms	92 - 337 MB	NPU
DDRNet23-Slim	ONNX	w8a8	Qualcomm® QCS6490	299.319 ms	198 - 215 MB	CPU
DDRNet23-Slim	ONNX	w8a8	Qualcomm® QCS8550 (Proxy)	57.348 ms	86 - 217 MB	NPU
DDRNet23-Slim	ONNX	w8a8	Qualcomm® QCS9075	65.461 ms	86 - 89 MB	NPU
DDRNet23-Slim	ONNX	w8a8	Qualcomm® QCM6690	266.19 ms	207 - 216 MB	CPU
DDRNet23-Slim	ONNX	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	42.08 ms	81 - 269 MB	NPU
DDRNet23-Slim	ONNX	w8a8	Snapdragon® 7 Gen 4 Mobile	250.665 ms	141 - 150 MB	CPU
DDRNet23-Slim	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	10.368 ms	12 - 248 MB	NPU
DDRNet23-Slim	QNN_DLC	float	Snapdragon® X2 Elite	11.788 ms	24 - 24 MB	NPU
DDRNet23-Slim	QNN_DLC	float	Snapdragon® X Elite	33.664 ms	24 - 24 MB	NPU
DDRNet23-Slim	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	22.364 ms	23 - 306 MB	NPU
DDRNet23-Slim	QNN_DLC	float	Qualcomm® QCS8275 (Proxy)	98.295 ms	24 - 222 MB	NPU
DDRNet23-Slim	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	32.594 ms	24 - 26 MB	NPU
DDRNet23-Slim	QNN_DLC	float	Qualcomm® SA8775P	40.269 ms	24 - 223 MB	NPU
DDRNet23-Slim	QNN_DLC	float	Qualcomm® QCS8450 (Proxy)	66.847 ms	5 - 290 MB	NPU
DDRNet23-Slim	QNN_DLC	float	Qualcomm® SA7255P	98.295 ms	24 - 222 MB	NPU
DDRNet23-Slim	QNN_DLC	float	Qualcomm® SA8295P	42.992 ms	24 - 230 MB	NPU
DDRNet23-Slim	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	15.419 ms	23 - 246 MB	NPU
DDRNet23-Slim	QNN_DLC	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	46.991 ms	6 - 238 MB	NPU
DDRNet23-Slim	QNN_DLC	w8a8	Snapdragon® X2 Elite	47.765 ms	6 - 6 MB	NPU
DDRNet23-Slim	QNN_DLC	w8a8	Snapdragon® X Elite	58.793 ms	6 - 6 MB	NPU
DDRNet23-Slim	QNN_DLC	w8a8	Snapdragon® 8 Gen 3 Mobile	41.852 ms	6 - 257 MB	NPU
DDRNet23-Slim	QNN_DLC	w8a8	Qualcomm® QCS8275 (Proxy)	107.874 ms	6 - 205 MB	NPU
DDRNet23-Slim	QNN_DLC	w8a8	Qualcomm® QCS8550 (Proxy)	56.064 ms	6 - 8 MB	NPU
DDRNet23-Slim	QNN_DLC	w8a8	Qualcomm® SA8775P	56.885 ms	6 - 205 MB	NPU
DDRNet23-Slim	QNN_DLC	w8a8	Qualcomm® QCS9075	60.444 ms	6 - 14 MB	NPU
DDRNet23-Slim	QNN_DLC	w8a8	Qualcomm® QCS8450 (Proxy)	61.535 ms	6 - 257 MB	NPU
DDRNet23-Slim	QNN_DLC	w8a8	Qualcomm® SA7255P	107.874 ms	6 - 205 MB	NPU
DDRNet23-Slim	QNN_DLC	w8a8	Qualcomm® SA8295P	64.37 ms	6 - 208 MB	NPU
DDRNet23-Slim	QNN_DLC	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	40.531 ms	6 - 222 MB	NPU
DDRNet23-Slim	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	10.336 ms	2 - 242 MB	NPU
DDRNet23-Slim	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	22.47 ms	2 - 294 MB	NPU
DDRNet23-Slim	TFLITE	float	Qualcomm® QCS8275 (Proxy)	98.31 ms	3 - 206 MB	NPU
DDRNet23-Slim	TFLITE	float	Qualcomm® QCS8550 (Proxy)	32.619 ms	2 - 5 MB	NPU
DDRNet23-Slim	TFLITE	float	Qualcomm® SA8775P	40.119 ms	2 - 206 MB	NPU
DDRNet23-Slim	TFLITE	float	Qualcomm® QCS9075	53.826 ms	0 - 41 MB	NPU
DDRNet23-Slim	TFLITE	float	Qualcomm® QCS8450 (Proxy)	68.162 ms	0 - 295 MB	NPU
DDRNet23-Slim	TFLITE	float	Qualcomm® SA7255P	98.31 ms	3 - 206 MB	NPU
DDRNet23-Slim	TFLITE	float	Qualcomm® SA8295P	43.006 ms	2 - 216 MB	NPU
DDRNet23-Slim	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	15.283 ms	2 - 228 MB	NPU
DDRNet23-Slim	TFLITE	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	44.914 ms	1 - 232 MB	NPU
DDRNet23-Slim	TFLITE	w8a8	Snapdragon® 8 Gen 3 Mobile	37.128 ms	1 - 253 MB	NPU
DDRNet23-Slim	TFLITE	w8a8	Qualcomm® QCS6490	200.523 ms	10 - 78 MB	NPU
DDRNet23-Slim	TFLITE	w8a8	Qualcomm® QCS8275 (Proxy)	95.166 ms	0 - 197 MB	NPU
DDRNet23-Slim	TFLITE	w8a8	Qualcomm® QCS8550 (Proxy)	49.099 ms	1 - 3 MB	NPU
DDRNet23-Slim	TFLITE	w8a8	Qualcomm® SA8775P	49.614 ms	1 - 198 MB	NPU
DDRNet23-Slim	TFLITE	w8a8	Qualcomm® QCS9075	51.926 ms	0 - 15 MB	NPU
DDRNet23-Slim	TFLITE	w8a8	Qualcomm® QCM6690	210.792 ms	9 - 195 MB	NPU
DDRNet23-Slim	TFLITE	w8a8	Qualcomm® QCS8450 (Proxy)	56.178 ms	1 - 253 MB	NPU
DDRNet23-Slim	TFLITE	w8a8	Qualcomm® SA7255P	95.166 ms	0 - 197 MB	NPU
DDRNet23-Slim	TFLITE	w8a8	Qualcomm® SA8295P	56.011 ms	1 - 203 MB	NPU
DDRNet23-Slim	TFLITE	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	65.077 ms	1 - 215 MB	NPU
DDRNet23-Slim	TFLITE	w8a8	Snapdragon® 7 Gen 4 Mobile	66.779 ms	11 - 211 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