BEVPoolv2: A Cutting-edge Implementation of BEVDet Toward Deployment

TL;DR

BEVPoolv2 significantly improves the efficiency and deployment flexibility of BEVDet by optimizing view transformation, reducing computation and storage, and enabling fast deployment on backends like TensorRT.

Contribution

The paper introduces BEVPoolv2, an optimized view transformation method that reduces computational and storage burdens, and demonstrates its effective deployment on various backends.

Findings

Processing time reduced to 0.82 ms at high resolution

Achieved 15.1x speedup over previous implementation

Scores 52.3 NDS on NuScenes validation set

Abstract

We release a new codebase version of the BEVDet, dubbed branch dev2.0. With dev2.0, we propose BEVPoolv2 upgrade the view transformation process from the perspective of engineering optimization, making it free from a huge burden in both calculation and storage aspects. It achieves this by omitting the calculation and preprocessing of the large frustum feature. As a result, it can be processed within 0.82 ms even with a large input resolution of 640x1600, which is 15.1 times the previous fastest implementation. Besides, it is also less cache consumptive when compared with the previous implementation, naturally as it no longer needs to store the large frustum feature. Last but not least, this also makes the deployment to the other backend handy. We offer an example of deployment to the TensorRT backend in branch dev2.0 and show how fast the BEVDet paradigm can be processed on it. Other than BEVPoolv2, we also select and integrate some substantial progress that was proposed in the past year. As an example configuration, BEVDet4D-R50-Depth-CBGS scores 52.3 NDS on the NuScenes validation set and can be processed at a speed of 16.4 FPS with the PyTorch backend. The code has been released to facilitate the study on https://github.com/HuangJunJie2017/BEVDet/tree/dev2.0.