BEVDet: Optimized for Qualcomm Devices

BEVDet is a machine learning model for generating a birds eye view represenation from the sensors(cameras) mounted on a vehicle.

This is based on the implementation of BEVDet 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	ONNX Runtime 1.24.3	Download
ONNX	w8a16_mixed_fp16	Universal	ONNX Runtime 1.24.3	Download
TFLITE	float	Universal		Download

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

Model Details

Model Type: Model_use_case.driver_assistance

Model Stats:

• Model checkpoint: bevdet-r50.pth
• Input resolution: 1 x 6 x 3 x 256 x 704
• Number of parameters: 44M
• Model size: 171 MB

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
BEVDet	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	1319.725 ms	252 - 262 MB	CPU
BEVDet	ONNX	float	Snapdragon® 8 Elite Mobile	1382.282 ms	247 - 259 MB	CPU
BEVDet	ONNX	float	Snapdragon® X2 Elite	615.254 ms	737 - 737 MB	CPU
BEVDet	ONNX	float	Snapdragon® X Elite	660.577 ms	732 - 732 MB	CPU
BEVDet	ONNX	float	Snapdragon® X Elite	660.577 ms	732 - 732 MB	CPU
BEVDet	ONNX	float	Snapdragon® 8 Gen 3 Mobile	2056.739 ms	213 - 222 MB	CPU
BEVDet	ONNX	float	Qualcomm® QCS8550 (Proxy)	2546.077 ms	187 - 189 MB	CPU
BEVDet	ONNX	float	Qualcomm® QCS9075	1523.186 ms	237 - 251 MB	CPU
BEVDet	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	1382.282 ms	247 - 259 MB	CPU
BEVDet	ONNX	w8a16_mixed_fp16	Snapdragon® 8 Elite Gen 5 Mobile	1946.927 ms	322 - 336 MB	CPU
BEVDet	ONNX	w8a16_mixed_fp16	Snapdragon® 8 Elite Mobile	1531.756 ms	325 - 338 MB	CPU
BEVDet	ONNX	w8a16_mixed_fp16	Snapdragon® X2 Elite	900.274 ms	709 - 709 MB	CPU
BEVDet	ONNX	w8a16_mixed_fp16	Snapdragon® X Elite	953.675 ms	1238 - 1238 MB	CPU
BEVDet	ONNX	w8a16_mixed_fp16	Snapdragon® X Elite	953.675 ms	1238 - 1238 MB	CPU
BEVDet	ONNX	w8a16_mixed_fp16	Snapdragon® 8 Gen 3 Mobile	2471.126 ms	360 - 376 MB	CPU
BEVDet	ONNX	w8a16_mixed_fp16	Qualcomm® QCS8550 (Proxy)	2660.632 ms	398 - 405 MB	CPU
BEVDet	ONNX	w8a16_mixed_fp16	Qualcomm® QCS9075	1843.167 ms	422 - 431 MB	CPU
BEVDet	ONNX	w8a16_mixed_fp16	Snapdragon® 8 Elite For Galaxy Mobile	1531.756 ms	325 - 338 MB	CPU
BEVDet	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	1166.896 ms	87 - 98 MB	CPU
BEVDet	TFLITE	float	Snapdragon® 8 Elite Mobile	1232.282 ms	108 - 121 MB	CPU
BEVDet	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	1791.251 ms	102 - 114 MB	CPU
BEVDet	TFLITE	float	Qualcomm® QCS8275 (Proxy)	3168.156 ms	128 - 137 MB	CPU
BEVDet	TFLITE	float	Qualcomm® QCS8550 (Proxy)	2071.447 ms	124 - 286 MB	CPU
BEVDet	TFLITE	float	Qualcomm® SA8775P	2488.685 ms	127 - 138 MB	CPU
BEVDet	TFLITE	float	Qualcomm® SA8775P	2488.685 ms	127 - 138 MB	CPU
BEVDet	TFLITE	float	Qualcomm® SA8775P	2488.685 ms	127 - 138 MB	CPU
BEVDet	TFLITE	float	Qualcomm® QCS9075	2390.859 ms	127 - 1330 MB	CPU
BEVDet	TFLITE	float	Qualcomm® QCS8450 (Proxy)	2752.156 ms	125 - 143 MB	CPU
BEVDet	TFLITE	float	Qualcomm® SA7255P	3168.156 ms	128 - 137 MB	CPU
BEVDet	TFLITE	float	Qualcomm® SA8295P	1835.492 ms	127 - 138 MB	CPU
BEVDet	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	1232.282 ms	108 - 121 MB	CPU

License

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

References

• BEVDet: High-Performance Multi-Camera 3D Object Detection in Bird-Eye-View
• 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.