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	QAIRT 2.42, ONNX Runtime 1.24.3	Download
ONNX	w8a16_mixed_fp16	Universal	QAIRT 2.42, ONNX Runtime 1.24.3	Download
TFLITE	float	Universal	QAIRT 2.43, TFLite 2.19.1	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	1294.705 ms	243 - 255 MB	CPU
BEVDet	ONNX	float	Snapdragon® X2 Elite	593.981 ms	735 - 735 MB	CPU
BEVDet	ONNX	float	Snapdragon® X Elite	716.192 ms	731 - 731 MB	CPU
BEVDet	ONNX	float	Snapdragon® 8 Gen 3 Mobile	2184.526 ms	191 - 201 MB	CPU
BEVDet	ONNX	float	Qualcomm® QCS8550 (Proxy)	2525.094 ms	175 - 190 MB	CPU
BEVDet	ONNX	float	Qualcomm® QCS9075	1518.109 ms	233 - 262 MB	CPU
BEVDet	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	1430.074 ms	239 - 248 MB	CPU
BEVDet	ONNX	w8a16_mixed_fp16	Snapdragon® 8 Elite Gen 5 Mobile	1952.078 ms	315 - 329 MB	CPU
BEVDet	ONNX	w8a16_mixed_fp16	Snapdragon® X2 Elite	767.638 ms	1230 - 1230 MB	CPU
BEVDet	ONNX	w8a16_mixed_fp16	Snapdragon® X Elite	939.559 ms	1238 - 1238 MB	CPU
BEVDet	ONNX	w8a16_mixed_fp16	Snapdragon® 8 Gen 3 Mobile	2334.431 ms	348 - 363 MB	CPU
BEVDet	ONNX	w8a16_mixed_fp16	Qualcomm® QCS8550 (Proxy)	2703.107 ms	383 - 402 MB	CPU
BEVDet	ONNX	w8a16_mixed_fp16	Qualcomm® QCS9075	1875.752 ms	412 - 432 MB	CPU
BEVDet	ONNX	w8a16_mixed_fp16	Snapdragon® 8 Elite For Galaxy Mobile	1626.633 ms	322 - 336 MB	CPU
BEVDet	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	1158.727 ms	87 - 98 MB	CPU
BEVDet	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	1723.799 ms	100 - 111 MB	CPU
BEVDet	TFLITE	float	Qualcomm® QCS8275 (Proxy)	3151.407 ms	128 - 137 MB	CPU
BEVDet	TFLITE	float	Qualcomm® QCS8550 (Proxy)	2076.946 ms	0 - 856 MB	CPU
BEVDet	TFLITE	float	Qualcomm® SA8775P	2477.676 ms	127 - 138 MB	CPU
BEVDet	TFLITE	float	Qualcomm® QCS9075	2397.198 ms	127 - 1331 MB	CPU
BEVDet	TFLITE	float	Qualcomm® QCS8450 (Proxy)	2413.945 ms	124 - 141 MB	CPU
BEVDet	TFLITE	float	Qualcomm® SA7255P	3151.407 ms	128 - 137 MB	CPU
BEVDet	TFLITE	float	Qualcomm® SA8295P	1830.266 ms	127 - 137 MB	CPU
BEVDet	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	1317.654 ms	136 - 149 MB	CPU

License

• The license for the original implementation of BEVDet can be found [here](https://github.com/HuangJunJie2017/BEVDet/blob/dev3.0/LICENSE https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/Qualcomm+AI+Hub+Proprietary+License.pdf).

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.