Point-Plane Projections for Accurate LiDAR Semantic Segmentation in Small Data Scenarios
This addresses the challenge of high computational complexity and data requirements in LiDAR segmentation for applications like autonomous driving, though it is incremental as it builds on point-based methods.
The paper tackles the problem of LiDAR semantic segmentation in data-scarce scenarios by using point-plane projections to learn features from 2D representations and a geometry-aware data augmentation technique, achieving significant improvements in limited-data settings and competitive results on standard datasets like SemanticKITTI and PandaSet.
LiDAR point cloud semantic segmentation is essential for interpreting 3D environments in applications such as autonomous driving and robotics. Recent methods achieve strong performance by exploiting different point cloud representations or incorporating data from other sensors, such as cameras or external datasets. However, these approaches often suffer from high computational complexity and require large amounts of training data, limiting their generalization in data-scarce scenarios. In this paper, we improve the performance of point-based methods by effectively learning features from 2D representations through point-plane projections, enabling the extraction of complementary information while relying solely on LiDAR data. Additionally, we introduce a geometry-aware technique for data augmentation that aligns with LiDAR sensor properties and mitigates class imbalance. We implemented and evaluated our method that applies point-plane projections onto multiple informative 2D representations of the point cloud. Experiments demonstrate that this approach leads to significant improvements in limited-data scenarios, while also achieving competitive results on two publicly available standard datasets, as SemanticKITTI and PandaSet. The code of our method is available at https://github.com/SiMoM0/3PNet