PointCLIMB: An Exemplar-Free Point Cloud Class Incremental Benchmark

TL;DR

PointCLIMB introduces a new benchmark for exemplar-free class incremental learning on 3D point clouds, addressing memory and legality concerns, and evaluates various backbones on the ModelNet40 dataset.

Contribution

This paper pioneers a benchmark for exemplar-free class incremental learning on 3D point clouds, focusing on practical scenarios with memory and legal constraints.

Findings

Demonstrates the effectiveness of different backbones on the benchmark

Provides insights into performance trade-offs in exemplar-free 3D continual learning

Establishes a new standard for future research in 3D class incremental learning

Abstract

Point clouds offer comprehensive and precise data regarding the contour and configuration of objects. Employing such geometric and topological 3D information of objects in class incremental learning can aid endless application in 3D-computer vision. Well known 3D-point cloud class incremental learning methods for addressing catastrophic forgetting generally entail the usage of previously encountered data, which can present difficulties in situations where there are restrictions on memory or when there are concerns about the legality of the data. Towards this we pioneer to leverage exemplar free class incremental learning on Point Clouds. In this paper we propose PointCLIMB: An exemplar Free Class Incremental Learning Benchmark. We focus on a pragmatic perspective to consider novel classes for class incremental learning on 3D point clouds. We setup a benchmark for 3D Exemplar free class incremental learning. We investigate performance of various backbones on 3D-Exemplar Free Class Incremental Learning framework. We demonstrate our results on ModelNet40 dataset.