Midas-V2: Optimized for Qualcomm Devices

Midas is designed for estimating depth at each point in an image.

This is based on the implementation of Midas-V2 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 1.25.0	Download
ONNX	w8a8	Universal	QAIRT 2.42, 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
TFLITE	w8a8	Universal	QAIRT 2.45	Download

For more device-specific assets and performance metrics, visit Midas-V2 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 Midas-V2 on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.depth_estimation

Model Stats:

• Model checkpoint: MiDaS_small
• Input resolution: 256x256
• Number of parameters: 16.6M
• Model size (float): 63.2 MB
• Model size (w8a8): 16.9 MB

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
Midas-V2	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	1.293 ms	0 - 165 MB	NPU
Midas-V2	ONNX	float	Snapdragon® X2 Elite	1.312 ms	211 - 211 MB	NPU
Midas-V2	ONNX	float	Snapdragon® X Elite	2.748 ms	148 - 148 MB	NPU
Midas-V2	ONNX	float	Snapdragon® 8 Gen 3 Mobile	1.923 ms	1 - 201 MB	NPU
Midas-V2	ONNX	float	Qualcomm® QCS8550 (Proxy)	2.766 ms	0 - 42 MB	NPU
Midas-V2	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	1.481 ms	0 - 163 MB	NPU
Midas-V2	ONNX	float	Qualcomm® QCS9075	3.87 ms	1 - 46 MB	NPU
Midas-V2	ONNX	float	Qualcomm® QCS8750	1.481 ms	0 - 163 MB	NPU
Midas-V2	ONNX	float	Qualcomm® QCS7181	2.748 ms	148 - 148 MB	NPU
Midas-V2	ONNX	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.634 ms	0 - 61 MB	NPU
Midas-V2	ONNX	w8a8	Snapdragon® X2 Elite	0.615 ms	165 - 165 MB	NPU
Midas-V2	ONNX	w8a8	Snapdragon® 8 Gen 3 Mobile	0.913 ms	0 - 99 MB	NPU
Midas-V2	ONNX	w8a8	Qualcomm® QCS8550 (Proxy)	1.32 ms	0 - 47 MB	NPU
Midas-V2	ONNX	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	0.742 ms	0 - 58 MB	NPU
Midas-V2	ONNX	w8a8	Qualcomm® QCS9075	1.465 ms	0 - 45 MB	NPU
Midas-V2	ONNX	w8a8	Qualcomm® QCS8750	0.742 ms	0 - 58 MB	NPU
Midas-V2	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	1.278 ms	1 - 164 MB	NPU
Midas-V2	QNN_DLC	float	Snapdragon® X2 Elite	1.589 ms	1 - 1 MB	NPU
Midas-V2	QNN_DLC	float	Snapdragon® X Elite	3.243 ms	1 - 1 MB	NPU
Midas-V2	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	2.126 ms	0 - 192 MB	NPU
Midas-V2	QNN_DLC	float	Qualcomm® QCS8275	12.103 ms	1 - 159 MB	NPU
Midas-V2	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	3.048 ms	0 - 8 MB	NPU
Midas-V2	QNN_DLC	float	Qualcomm® SA8775P	4.243 ms	0 - 159 MB	NPU
Midas-V2	QNN_DLC	float	Qualcomm® SA8650P	4.243 ms	0 - 159 MB	NPU
Midas-V2	QNN_DLC	float	Qualcomm® SA8255P	4.243 ms	0 - 159 MB	NPU
Midas-V2	QNN_DLC	float	Qualcomm® QCS8450 (Proxy)	6.82 ms	0 - 192 MB	NPU
Midas-V2	QNN_DLC	float	Qualcomm® SA8295P	5.47 ms	1 - 154 MB	NPU
Midas-V2	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	1.601 ms	0 - 160 MB	NPU
Midas-V2	QNN_DLC	float	Qualcomm® SA7255P	12.103 ms	1 - 159 MB	NPU
Midas-V2	QNN_DLC	float	Qualcomm® QCS9075	4.461 ms	1 - 3 MB	NPU
Midas-V2	QNN_DLC	float	Qualcomm® QCS8750	1.601 ms	0 - 160 MB	NPU
Midas-V2	QNN_DLC	float	Qualcomm® QCS7181	3.243 ms	1 - 1 MB	NPU
Midas-V2	QNN_DLC	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.569 ms	0 - 50 MB	NPU
Midas-V2	QNN_DLC	w8a8	Snapdragon® X2 Elite	0.739 ms	0 - 0 MB	NPU
Midas-V2	QNN_DLC	w8a8	Snapdragon® X Elite	1.441 ms	0 - 0 MB	NPU
Midas-V2	QNN_DLC	w8a8	Snapdragon® 8 Gen 3 Mobile	0.918 ms	0 - 81 MB	NPU
Midas-V2	QNN_DLC	w8a8	Qualcomm® QCS6490	4.082 ms	0 - 2 MB	NPU
Midas-V2	QNN_DLC	w8a8	Qualcomm® QCS8275	2.867 ms	0 - 47 MB	NPU
Midas-V2	QNN_DLC	w8a8	Qualcomm® QCS8550 (Proxy)	1.319 ms	0 - 11 MB	NPU
Midas-V2	QNN_DLC	w8a8	Qualcomm® SA8775P	1.571 ms	0 - 52 MB	NPU
Midas-V2	QNN_DLC	w8a8	Qualcomm® SA8650P	1.571 ms	0 - 52 MB	NPU
Midas-V2	QNN_DLC	w8a8	Qualcomm® SA8255P	1.571 ms	0 - 52 MB	NPU
Midas-V2	QNN_DLC	w8a8	Qualcomm® SA7255P	2.867 ms	0 - 47 MB	NPU
Midas-V2	QNN_DLC	w8a8	Qualcomm® SA8295P	2.224 ms	0 - 47 MB	NPU
Midas-V2	QNN_DLC	w8a8	Snapdragon® 7 Gen 4 Mobile	1.584 ms	0 - 172 MB	NPU
Midas-V2	QNN_DLC	w8a8	Qualcomm® QCM6690	9.056 ms	0 - 173 MB	NPU
Midas-V2	QNN_DLC	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	0.659 ms	0 - 50 MB	NPU
Midas-V2	QNN_DLC	w8a8	Qualcomm® QCS9075	1.445 ms	0 - 2 MB	NPU
Midas-V2	QNN_DLC	w8a8	Qualcomm® QCS8450 (Proxy)	1.856 ms	0 - 85 MB	NPU
Midas-V2	QNN_DLC	w8a8	Qualcomm® QCS7790	1.584 ms	0 - 172 MB	NPU
Midas-V2	QNN_DLC	w8a8	Qualcomm® QCS8750	0.659 ms	0 - 50 MB	NPU
Midas-V2	QNN_DLC	w8a8	Qualcomm® QCS7181	1.441 ms	0 - 0 MB	NPU
Midas-V2	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	1.279 ms	0 - 174 MB	NPU
Midas-V2	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	2.123 ms	0 - 227 MB	NPU
Midas-V2	TFLITE	float	Qualcomm® QCS8275	12.116 ms	0 - 173 MB	NPU
Midas-V2	TFLITE	float	Qualcomm® QCS8550 (Proxy)	3.013 ms	0 - 3 MB	NPU
Midas-V2	TFLITE	float	Qualcomm® SA8775P	4.282 ms	0 - 174 MB	NPU
Midas-V2	TFLITE	float	Qualcomm® SA8650P	4.282 ms	0 - 174 MB	NPU
Midas-V2	TFLITE	float	Qualcomm® SA8255P	4.282 ms	0 - 174 MB	NPU
Midas-V2	TFLITE	float	Qualcomm® QCS8450 (Proxy)	6.841 ms	0 - 214 MB	NPU
Midas-V2	TFLITE	float	Qualcomm® SA8295P	5.467 ms	0 - 155 MB	NPU
Midas-V2	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	1.606 ms	0 - 171 MB	NPU
Midas-V2	TFLITE	float	Qualcomm® SA7255P	12.116 ms	0 - 173 MB	NPU
Midas-V2	TFLITE	float	Qualcomm® QCS9075	4.386 ms	0 - 39 MB	NPU
Midas-V2	TFLITE	float	Qualcomm® QCS8750	1.606 ms	0 - 171 MB	NPU
Midas-V2	TFLITE	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.491 ms	0 - 48 MB	NPU
Midas-V2	TFLITE	w8a8	Snapdragon® 8 Gen 3 Mobile	0.749 ms	0 - 78 MB	NPU
Midas-V2	TFLITE	w8a8	Qualcomm® QCS6490	3.614 ms	0 - 28 MB	NPU
Midas-V2	TFLITE	w8a8	Qualcomm® QCS8275	2.49 ms	0 - 44 MB	NPU
Midas-V2	TFLITE	w8a8	Qualcomm® QCS8550 (Proxy)	1.079 ms	0 - 36 MB	NPU
Midas-V2	TFLITE	w8a8	Qualcomm® SA8775P	1.352 ms	0 - 48 MB	NPU
Midas-V2	TFLITE	w8a8	Qualcomm® SA8650P	1.352 ms	0 - 48 MB	NPU
Midas-V2	TFLITE	w8a8	Qualcomm® SA8255P	1.352 ms	0 - 48 MB	NPU
Midas-V2	TFLITE	w8a8	Qualcomm® SA7255P	2.49 ms	0 - 44 MB	NPU
Midas-V2	TFLITE	w8a8	Qualcomm® SA8295P	1.952 ms	0 - 43 MB	NPU
Midas-V2	TFLITE	w8a8	Snapdragon® 7 Gen 4 Mobile	1.34 ms	0 - 167 MB	NPU
Midas-V2	TFLITE	w8a8	Qualcomm® QCM6690	8.343 ms	0 - 171 MB	NPU
Midas-V2	TFLITE	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	0.555 ms	0 - 46 MB	NPU
Midas-V2	TFLITE	w8a8	Qualcomm® QCS9075	1.174 ms	0 - 20 MB	NPU
Midas-V2	TFLITE	w8a8	Qualcomm® QCS8450 (Proxy)	1.596 ms	0 - 77 MB	NPU
Midas-V2	TFLITE	w8a8	Qualcomm® QCS7790	1.34 ms	0 - 167 MB	NPU
Midas-V2	TFLITE	w8a8	Qualcomm® QCS8750	0.555 ms	0 - 46 MB	NPU

License

• The license for the original implementation of Midas-V2 can be found here.

References

• Towards Robust Monocular Depth Estimation: Mixing Datasets for Zero-shot Cross-dataset Transfer
• 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.