Continual Learning for Pose-Agnostic Object Recognition in 3D Point Clouds
This work addresses a practical limitation in continual learning for 3D vision, making it more applicable to real-world scenarios where object poses vary, though it is incremental as it builds on existing continual learning frameworks.
The paper tackles the problem of continual learning for 3D point cloud recognition when object poses are dynamic and unpredictable, proposing a method that injects equivariance as prior knowledge to distill geometric information, achieving improved performance in pose-agnostic scenarios on mainstream datasets.
Continual Learning aims to learn multiple incoming new tasks continually, and to keep the performance of learned tasks at a consistent level. However, existing research on continual learning assumes the pose of the object is pre-defined and well-aligned. For practical application, this work focuses on pose-agnostic continual learning tasks, where the object's pose changes dynamically and unpredictably. The point cloud augmentation adopted from past approaches would sharply rise with the task increment in the continual learning process. To address this problem, we inject the equivariance as the additional prior knowledge into the networks. We proposed a novel continual learning model that effectively distillates previous tasks' geometric equivariance information. The experiments show that our method overcomes the challenge of pose-agnostic scenarios in several mainstream point cloud datasets. We further conduct ablation studies to evaluate the validation of each component of our approach.