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 1.24.1	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 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® X Elite	2.937 ms	35 - 35 MB	NPU
Midas-V2	ONNX	float	Snapdragon® 8 Gen 3 Mobile	1.92 ms	1 - 201 MB	NPU
Midas-V2	ONNX	float	Qualcomm® QCS8550 (Proxy)	2.736 ms	0 - 7 MB	NPU
Midas-V2	ONNX	float	Qualcomm® QCS9075	3.873 ms	1 - 4 MB	NPU
Midas-V2	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	1.484 ms	0 - 160 MB	NPU
Midas-V2	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	1.294 ms	0 - 164 MB	NPU
Midas-V2	ONNX	float	Snapdragon® X2 Elite	1.314 ms	35 - 35 MB	NPU
Midas-V2	ONNX	w8a8	Snapdragon® X Elite	1.51 ms	18 - 18 MB	NPU
Midas-V2	ONNX	w8a8	Snapdragon® 8 Gen 3 Mobile	0.926 ms	0 - 94 MB	NPU
Midas-V2	ONNX	w8a8	Qualcomm® QCS8550 (Proxy)	1.373 ms	0 - 4 MB	NPU
Midas-V2	ONNX	w8a8	Qualcomm® QCS9075	1.506 ms	0 - 3 MB	NPU
Midas-V2	ONNX	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	0.758 ms	0 - 55 MB	NPU
Midas-V2	ONNX	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.648 ms	0 - 58 MB	NPU
Midas-V2	ONNX	w8a8	Snapdragon® X2 Elite	0.638 ms	20 - 20 MB	NPU
Midas-V2	QNN_DLC	float	Snapdragon® X Elite	3.296 ms	1 - 1 MB	NPU
Midas-V2	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	2.127 ms	0 - 195 MB	NPU
Midas-V2	QNN_DLC	float	Qualcomm® QCS8275 (Proxy)	12.117 ms	1 - 159 MB	NPU
Midas-V2	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	3.063 ms	1 - 3 MB	NPU
Midas-V2	QNN_DLC	float	Qualcomm® SA8775P	4.248 ms	1 - 161 MB	NPU
Midas-V2	QNN_DLC	float	Qualcomm® QCS9075	4.492 ms	1 - 3 MB	NPU
Midas-V2	QNN_DLC	float	Qualcomm® QCS8450 (Proxy)	6.827 ms	0 - 192 MB	NPU
Midas-V2	QNN_DLC	float	Qualcomm® SA7255P	12.117 ms	1 - 159 MB	NPU
Midas-V2	QNN_DLC	float	Qualcomm® SA8295P	5.506 ms	1 - 161 MB	NPU
Midas-V2	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	1.616 ms	1 - 161 MB	NPU
Midas-V2	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	1.274 ms	1 - 166 MB	NPU
Midas-V2	QNN_DLC	float	Snapdragon® X2 Elite	1.589 ms	1 - 1 MB	NPU
Midas-V2	QNN_DLC	w8a8	Snapdragon® X Elite	1.481 ms	0 - 0 MB	NPU
Midas-V2	QNN_DLC	w8a8	Snapdragon® 8 Gen 3 Mobile	0.921 ms	0 - 85 MB	NPU
Midas-V2	QNN_DLC	w8a8	Qualcomm® QCS6490	4.35 ms	0 - 2 MB	NPU
Midas-V2	QNN_DLC	w8a8	Qualcomm® QCS8275 (Proxy)	2.852 ms	0 - 46 MB	NPU
Midas-V2	QNN_DLC	w8a8	Qualcomm® QCS8550 (Proxy)	1.327 ms	0 - 2 MB	NPU
Midas-V2	QNN_DLC	w8a8	Qualcomm® SA8775P	1.599 ms	0 - 50 MB	NPU
Midas-V2	QNN_DLC	w8a8	Qualcomm® QCS9075	1.46 ms	2 - 4 MB	NPU
Midas-V2	QNN_DLC	w8a8	Qualcomm® QCM6690	9.001 ms	0 - 173 MB	NPU
Midas-V2	QNN_DLC	w8a8	Qualcomm® QCS8450 (Proxy)	1.855 ms	0 - 81 MB	NPU
Midas-V2	QNN_DLC	w8a8	Qualcomm® SA7255P	2.852 ms	0 - 46 MB	NPU
Midas-V2	QNN_DLC	w8a8	Qualcomm® SA8295P	2.278 ms	0 - 45 MB	NPU
Midas-V2	QNN_DLC	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	0.683 ms	0 - 46 MB	NPU
Midas-V2	QNN_DLC	w8a8	Snapdragon® 7 Gen 4 Mobile	1.591 ms	0 - 172 MB	NPU
Midas-V2	QNN_DLC	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.555 ms	0 - 50 MB	NPU
Midas-V2	QNN_DLC	w8a8	Snapdragon® X2 Elite	0.73 ms	0 - 0 MB	NPU
Midas-V2	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	2.124 ms	0 - 232 MB	NPU
Midas-V2	TFLITE	float	Qualcomm® QCS8275 (Proxy)	12.164 ms	0 - 179 MB	NPU
Midas-V2	TFLITE	float	Qualcomm® QCS8550 (Proxy)	3.048 ms	0 - 2 MB	NPU
Midas-V2	TFLITE	float	Qualcomm® SA8775P	4.313 ms	0 - 179 MB	NPU
Midas-V2	TFLITE	float	Qualcomm® QCS9075	4.441 ms	0 - 39 MB	NPU
Midas-V2	TFLITE	float	Qualcomm® QCS8450 (Proxy)	6.815 ms	0 - 218 MB	NPU
Midas-V2	TFLITE	float	Qualcomm® SA7255P	12.164 ms	0 - 179 MB	NPU
Midas-V2	TFLITE	float	Qualcomm® SA8295P	5.52 ms	0 - 159 MB	NPU
Midas-V2	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	1.637 ms	0 - 182 MB	NPU
Midas-V2	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	1.273 ms	0 - 179 MB	NPU
Midas-V2	TFLITE	w8a8	Snapdragon® 8 Gen 3 Mobile	0.758 ms	0 - 78 MB	NPU
Midas-V2	TFLITE	w8a8	Qualcomm® QCS6490	3.522 ms	0 - 28 MB	NPU
Midas-V2	TFLITE	w8a8	Qualcomm® QCS8275 (Proxy)	2.455 ms	0 - 44 MB	NPU
Midas-V2	TFLITE	w8a8	Qualcomm® QCS8550 (Proxy)	1.058 ms	0 - 12 MB	NPU
Midas-V2	TFLITE	w8a8	Qualcomm® SA8775P	1.373 ms	0 - 48 MB	NPU
Midas-V2	TFLITE	w8a8	Qualcomm® QCS9075	1.212 ms	0 - 20 MB	NPU
Midas-V2	TFLITE	w8a8	Qualcomm® QCM6690	8.324 ms	0 - 171 MB	NPU
Midas-V2	TFLITE	w8a8	Qualcomm® QCS8450 (Proxy)	1.59 ms	0 - 78 MB	NPU
Midas-V2	TFLITE	w8a8	Qualcomm® SA7255P	2.455 ms	0 - 44 MB	NPU
Midas-V2	TFLITE	w8a8	Qualcomm® SA8295P	2.005 ms	0 - 43 MB	NPU
Midas-V2	TFLITE	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	0.596 ms	0 - 50 MB	NPU
Midas-V2	TFLITE	w8a8	Snapdragon® 7 Gen 4 Mobile	1.346 ms	0 - 167 MB	NPU
Midas-V2	TFLITE	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.489 ms	0 - 49 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.