PSA-SSL: Pose and Size-aware Self-Supervised Learning on LiDAR Point Clouds
This work addresses the need for better geometric feature learning in 3D perception for autonomous driving, offering a novel SSL method that enhances performance on tasks like segmentation and detection with limited labeled data.
The paper tackles the problem of self-supervised learning on LiDAR point clouds by proposing PSA-SSL, which retains geometric information like object pose and scale through a bounding box regression pretext task and LiDAR beam pattern augmentation, achieving significant improvements in 3D semantic segmentation and object detection with up to 10 times fewer labels.
Self-supervised learning (SSL) on 3D point clouds has the potential to learn feature representations that can transfer to diverse sensors and multiple downstream perception tasks. However, recent SSL approaches fail to define pretext tasks that retain geometric information such as object pose and scale, which can be detrimental to the performance of downstream localization and geometry-sensitive 3D scene understanding tasks, such as 3D semantic segmentation and 3D object detection. We propose PSA-SSL, a novel extension to point cloud SSL that learns object pose and size-aware (PSA) features. Our approach defines a self-supervised bounding box regression pretext task, which retains object pose and size information. Furthermore, we incorporate LiDAR beam pattern augmentation on input point clouds, which encourages learning sensor-agnostic features. Our experiments demonstrate that with a single pretrained model, our light-weight yet effective extensions achieve significant improvements on 3D semantic segmentation with limited labels across popular autonomous driving datasets (Waymo, nuScenes, SemanticKITTI). Moreover, our approach outperforms other state-of-the-art SSL methods on 3D semantic segmentation (using up to 10 times less labels), as well as on 3D object detection. Our code will be released on https://github.com/TRAILab/PSA-SSL.