CVT: Optimized for Qualcomm Devices

Cross-View Transformer generates real-time bird's-eye view maps from multiple vehicle cameras for autonomous driving.

This is based on the implementation of CVT 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	w8a16_mixed_fp16	Universal	QAIRT 2.45, ONNX Runtime 1.25.0	Download
QNN_DLC	float	Universal	QAIRT 2.45	Download
TFLITE	float	Universal	QAIRT 2.45	Download

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

Model Details

Model Type: Model_use_case.driver_assistance

Model Stats:

• Model checkpoint: vehicles_50k.pt
• Inference latency: RealTime
• Input resolution: 1x6x3x224x480
• Number of parameters: 1.33M
• Model size (float): 5.18 MB

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
CVT	ONNX	float	Snapdragon® X2 Elite	182.294 ms	8 - 8 MB	NPU
CVT	ONNX	float	Snapdragon® X Elite	270.902 ms	19 - 19 MB	NPU
CVT	ONNX	float	Snapdragon® 8 Gen 3 Mobile	245.125 ms	8 - 2683 MB	NPU
CVT	ONNX	float	Snapdragon® 8 Gen 1 Mobile	436.757 ms	8 - 2778 MB	NPU
CVT	ONNX	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	329.711 ms	0 - 25 MB	NPU
CVT	ONNX	float	Qualcomm® QCS8450	436.757 ms	8 - 2778 MB	NPU
CVT	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	169.577 ms	3 - 2415 MB	NPU
CVT	ONNX	float	Qualcomm® Dragonwing™ IQ-9075	321.618 ms	7 - 53 MB	NPU
CVT	ONNX	float	Snapdragon® 8 Elite Mobile	186.289 ms	3 - 2380 MB	NPU
CVT	ONNX	float	Qualcomm® Dragonwing™ Q-8750	186.289 ms	3 - 2380 MB	NPU
CVT	ONNX	float	Qualcomm® Dragonwing™ IQ-X7181	270.902 ms	19 - 19 MB	NPU
CVT	ONNX	w8a16_mixed_fp16	Snapdragon® 8 Gen 3 Mobile	263.712 ms	15 - 2659 MB	NPU
CVT	ONNX	w8a16_mixed_fp16	Qualcomm® Dragonwing™ QCS8550 (Proxy)	363.316 ms	0 - 28 MB	NPU
CVT	ONNX	w8a16_mixed_fp16	Qualcomm® Dragonwing™ IQ-9075	320.828 ms	13 - 16 MB	NPU
CVT	ONNX	w8a16_mixed_fp16	Snapdragon® 8 Elite Gen 5 Mobile	197.898 ms	12 - 2511 MB	NPU
CVT	ONNX	w8a16_mixed_fp16	Snapdragon® 8 Elite Mobile	216.688 ms	12 - 2416 MB	NPU
CVT	ONNX	w8a16_mixed_fp16	Qualcomm® Dragonwing™ Q-8750	216.688 ms	12 - 2416 MB	NPU
CVT	QNN_DLC	float	Snapdragon® X2 Elite	182.523 ms	7 - 7 MB	NPU
CVT	QNN_DLC	float	Snapdragon® X Elite	286.686 ms	7 - 7 MB	NPU
CVT	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	220.366 ms	7 - 2575 MB	NPU
CVT	QNN_DLC	float	Snapdragon® 8 Gen 1 Mobile	440.531 ms	7 - 2720 MB	NPU
CVT	QNN_DLC	float	Qualcomm® QCS8275	425.082 ms	0 - 2212 MB	NPU
CVT	QNN_DLC	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	332.747 ms	8 - 12 MB	NPU
CVT	QNN_DLC	float	Qualcomm® SA8775P	292.829 ms	2 - 2213 MB	NPU
CVT	QNN_DLC	float	Qualcomm® SA8650P	292.829 ms	2 - 2213 MB	NPU
CVT	QNN_DLC	float	Qualcomm® SA8255P	292.829 ms	2 - 2213 MB	NPU
CVT	QNN_DLC	float	Qualcomm® QCS8450	440.531 ms	7 - 2720 MB	NPU
CVT	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	167.886 ms	7 - 2318 MB	NPU
CVT	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-9075	363.338 ms	7 - 17 MB	NPU
CVT	QNN_DLC	float	Qualcomm® SA7255P	425.082 ms	0 - 2212 MB	NPU
CVT	QNN_DLC	float	Snapdragon® 8 Elite Mobile	186.105 ms	7 - 2299 MB	NPU
CVT	QNN_DLC	float	Qualcomm® SA8295P	334.575 ms	0 - 2324 MB	NPU
CVT	QNN_DLC	float	Qualcomm® Dragonwing™ Q-8750	186.105 ms	7 - 2299 MB	NPU
CVT	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-X7181	286.686 ms	7 - 7 MB	NPU
CVT	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	226.281 ms	0 - 2673 MB	NPU
CVT	TFLITE	float	Snapdragon® 8 Gen 1 Mobile	389.924 ms	0 - 2726 MB	NPU
CVT	TFLITE	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	306.789 ms	0 - 3 MB	NPU
CVT	TFLITE	float	Qualcomm® SA8775P	572.418 ms	48 - 60 MB	CPU
CVT	TFLITE	float	Qualcomm® SA8650P	572.418 ms	48 - 60 MB	CPU
CVT	TFLITE	float	Qualcomm® SA8255P	572.418 ms	48 - 60 MB	CPU
CVT	TFLITE	float	Qualcomm® QCS8450	389.924 ms	0 - 2726 MB	NPU
CVT	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	166.0 ms	0 - 2314 MB	NPU
CVT	TFLITE	float	Qualcomm® Dragonwing™ IQ-9075	317.919 ms	0 - 36 MB	NPU
CVT	TFLITE	float	Snapdragon® 8 Elite Mobile	183.486 ms	1 - 2364 MB	NPU
CVT	TFLITE	float	Qualcomm® SA8295P	331.769 ms	0 - 2306 MB	NPU
CVT	TFLITE	float	Qualcomm® Dragonwing™ Q-8750	183.486 ms	1 - 2364 MB	NPU

License

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

References

• Cross-view Transformers for real-time Map-view Semantic Segmentation
• 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.