Vision-based Uneven BEV Representation Learning with Polar Rasterization and Surface Estimation

Zhi Liu, Shaoyu Chen, Xiaojie Guo, Xinggang Wang, Tianheng Cheng, Hongmei Zhu, Qian Zhang, Wenyu Liu, Yi Zhang

Student First Author: yes

Keywords: Polar Rasterization, Surface Estimation, BEV Segmentation

TL;DR: We propose PolarBEV for vision-based uneven BEV representation learning.

Abstract: In this work, we propose PolarBEV for vision-based uneven BEV representation learning. To adapt to the foreshortening effect of camera imaging, we rasterize the BEV space both angularly and radially, and introduce polar embedding decomposition to model the associations among polar grids. Polar grids are rearranged to an array-like regular representation for efficient processing. Besides, to determine the 2D-to-3D correspondence, we iteratively update the BEV surface based on a hypothetical plane, and adopt height-based feature transformation. PolarBEV keeps real-time inference speed on a single 2080Ti GPU, and outperforms other methods for both BEV semantic segmentation and BEV instance segmentation. Thorough ablations are presented to validate the design. The code will be released for facilitating further research.

Community Implementations: [ 1 code implementation](https://www.catalyzex.com/paper/arxiv:2207.01878/code)

Website: https://arxiv.org/pdf/2207.01878.pdf

Code: https://github.com/SuperZ-Liu/PolarBEV