StereoNet: Optimized for Qualcomm Devices

StereoNet is an end-to-end deep architecture for real-time stereo matching that produces high-quality, edge-preserved disparity maps from a rectified stereo image pair.

This is based on the implementation of StereoNet 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 1.27.1	Download
QNN_DLC	float	Universal	QAIRT 2.50	Download
TFLITE	float	Universal	QAIRT 2.50	Download

For more device-specific assets and performance metrics, visit StereoNet 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 StereoNet on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.depth_estimation

Model Stats:

• Input resolution: 786x490
• Model checkpoint: KeystoneDepth (epoch=21-step=696366.ckpt)
• Model size (float): 7.41 MB
• Number of parameters: 1.94M

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
StereoNet	ONNX	float	Snapdragon® 8 Elite Gen 5 For Galaxy Mobile	203.063 ms	5 - 2837 MB	NPU
StereoNet	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	243.626 ms	3 - 2784 MB	NPU
StereoNet	ONNX	float	Snapdragon® X2 Elite	208.952 ms	3 - 3 MB	NPU
StereoNet	ONNX	float	Snapdragon® X Elite	363.091 ms	40 - 40 MB	NPU
StereoNet	ONNX	float	Qualcomm® Dragonwing™ IQ-8275	474.054 ms	3 - 10 MB	NPU
StereoNet	ONNX	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	430.153 ms	0 - 49 MB	NPU
StereoNet	ONNX	float	Qualcomm® Dragonwing™ IQ-9075	443.394 ms	3 - 9 MB	NPU
StereoNet	ONNX	float	Qualcomm® Dragonwing™ IQ-X7181	363.091 ms	40 - 40 MB	NPU
StereoNet	ONNX	float	Qualcomm® Dragonwing™ Q-8750	243.626 ms	3 - 2784 MB	NPU
StereoNet	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 For Galaxy Mobile	186.254 ms	3 - 2846 MB	NPU
StereoNet	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	230.332 ms	0 - 2783 MB	NPU
StereoNet	QNN_DLC	float	Snapdragon® X2 Elite	185.521 ms	3 - 3 MB	NPU
StereoNet	QNN_DLC	float	Snapdragon® X Elite	346.697 ms	3 - 3 MB	NPU
StereoNet	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	266.448 ms	3 - 3861 MB	NPU
StereoNet	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-8275	457.901 ms	3 - 10 MB	NPU
StereoNet	QNN_DLC	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	434.369 ms	3 - 8 MB	NPU
StereoNet	QNN_DLC	float	Qualcomm® SA8650P	461.966 ms	1 - 3260 MB	NPU
StereoNet	QNN_DLC	float	Qualcomm® SA8255P	461.966 ms	1 - 3260 MB	NPU
StereoNet	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-9075	424.793 ms	5 - 11 MB	NPU
StereoNet	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-X7181	346.697 ms	3 - 3 MB	NPU
StereoNet	QNN_DLC	float	Qualcomm® Dragonwing™ Q-8750	230.332 ms	0 - 2783 MB	NPU
StereoNet	QNN_DLC	float	Qualcomm® SA8295P	506.741 ms	0 - 2944 MB	NPU
StereoNet	TFLITE	float	Snapdragon® 8 Elite Gen 5 For Galaxy Mobile	267.481 ms	25 - 3259 MB	NPU
StereoNet	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	282.473 ms	73 - 3218 MB	NPU
StereoNet	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	350.839 ms	72 - 4241 MB	NPU
StereoNet	TFLITE	float	Qualcomm® Dragonwing™ IQ-8275	576.12 ms	74 - 199 MB	NPU
StereoNet	TFLITE	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	576.624 ms	74 - 77 MB	NPU
StereoNet	TFLITE	float	Qualcomm® Dragonwing™ IQ-9075	557.185 ms	72 - 196 MB	NPU
StereoNet	TFLITE	float	Qualcomm® Dragonwing™ Q-8750	282.473 ms	73 - 3218 MB	NPU
StereoNet	TFLITE	float	Qualcomm® SA8295P	615.918 ms	74 - 3210 MB	NPU

License

• The license for the original implementation of StereoNet can be found here.

References

• StereoNet: Guided Hierarchical Refinement for Real-Time Edge-Aware Depth Prediction
• Source Model Implementation

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.