Addressing Data Misalignment in Image-LiDAR Fusion on Point Cloud Segmentation
This addresses a critical challenge in autonomous driving perception by improving fusion accuracy for better segmentation, but it appears incremental as it builds on existing SOTA models like 2DPASS.
The paper tackles the problem of data misalignment between cameras and LiDAR sensors in multi-sensor fusion models for autonomous driving, specifically focusing on point cloud segmentation, and proposes potential improvements to address this issue.
With the advent of advanced multi-sensor fusion models, there has been a notable enhancement in the performance of perception tasks within in terms of autonomous driving. Despite these advancements, the challenges persist, particularly in the fusion of data from cameras and LiDAR sensors. A critial concern is the accurate alignment of data from these disparate sensors. Our observations indicate that the projected positions of LiDAR points often misalign on the corresponding image. Furthermore, fusion models appear to struggle in accurately segmenting these misaligned points. In this paper, we would like to address this problem carefully, with a specific focus on the nuScenes dataset and the SOTA of fusion models 2DPASS, and providing the possible solutions or potential improvements.