Balanced Residual Distillation Learning for 3D Point Cloud Class-Incremental Semantic Segmentation

TL;DR

This paper introduces BRDL, a novel framework for class-incremental learning in 3D point cloud segmentation, effectively balancing old and new class knowledge to improve continual learning performance.

Contribution

It proposes a residual distillation and balanced pseudo-label strategy to enhance class-incremental learning, addressing class bias and knowledge refinement challenges.

Findings

Sets a new benchmark in 3D point cloud segmentation

Demonstrates superior balance between old and new class learning

Achieves state-of-the-art results in class-biased scenarios

Abstract

Class-incremental learning (CIL) enables continuous learning of new classes while mitigating catastrophic forgetting of old ones. For the performance breakthrough of CIL, it is essential yet challenging to effectively refine past knowledge from the base model and balance it with new learning. However, such a challenge has not been considered in current research. This work proposes a balanced residual distillation learning framework (BRDL) to address this gap and advance CIL performance. BRDL introduces a residual distillation strategy to dynamically refine past knowledge by expanding the network structure and a balanced pseudo-label learning strategy to mitigate class bias and balance learning between old and new classes. We apply the proposed BRDL to a challenging 3D point cloud semantic segmentation task where the data is unordered and unstructured. Extensive experimental results demonstrate that BRDL sets a new benchmark with an outstanding balance capability in class-biased scenarios.