Towards Cross-Platform Generalization: Domain Adaptive 3D Detection with Augmentation and Pseudo-Labeling
This work addresses domain adaptation in 3D detection for autonomous driving, but it is incremental as it builds on an existing framework with specific enhancements.
The paper tackled cross-platform generalization for 3D object detection by narrowing domain gaps with tailored data augmentation and pseudo-labeling, achieving a 3D AP of 62.67% for Car in phase-1 and up to 58.76% for Car in phase-2.
This technical report represents the award-winning solution to the Cross-platform 3D Object Detection task in the RoboSense2025 Challenge. Our approach is built upon PVRCNN++, an efficient 3D object detection framework that effectively integrates point-based and voxel-based features. On top of this foundation, we improve cross-platform generalization by narrowing domain gaps through tailored data augmentation and a self-training strategy with pseudo-labels. These enhancements enabled our approach to secure the 3rd place in the challenge, achieving a 3D AP of 62.67% for the Car category on the phase-1 target domain, and 58.76% and 49.81% for Car and Pedestrian categories respectively on the phase-2 target domain.