The 2nd Place Solution from the 3D Semantic Segmentation Track in the 2024 Waymo Open Dataset Challenge
This work addresses the need for improved perception in autonomous vehicles, but it is incremental as it builds on existing methods with data augmentation enhancements.
The authors tackled the problem of limited training diversity in LiDAR-based 3D semantic segmentation by introducing MixSeg3D, which combines a point cloud segmentation model with advanced data mixing strategies, achieving 2nd place in the 2024 Waymo Open Dataset Challenge.
3D semantic segmentation is one of the most crucial tasks in driving perception. The ability of a learning-based model to accurately perceive dense 3D surroundings often ensures the safe operation of autonomous vehicles. However, existing LiDAR-based 3D semantic segmentation databases consist of sequentially acquired LiDAR scans that are long-tailed and lack training diversity. In this report, we introduce MixSeg3D, a sophisticated combination of the strong point cloud segmentation model with advanced 3D data mixing strategies. Specifically, our approach integrates the MinkUNet family with LaserMix and PolarMix, two scene-scale data augmentation methods that blend LiDAR point clouds along the ego-scene's inclination and azimuth directions. Through empirical experiments, we demonstrate the superiority of MixSeg3D over the baseline and prior arts. Our team achieved 2nd place in the 3D semantic segmentation track of the 2024 Waymo Open Dataset Challenge.