ItTakesTwo: Leveraging Peer Representations for Semi-supervised LiDAR Semantic Segmentation

TL;DR

This paper introduces ItTakesTwo, a semi-supervised LiDAR segmentation framework that leverages peer representations and improved contrastive learning to significantly enhance performance over existing methods.

Contribution

ItTakesTwo is a novel semi-supervised framework that uses peer representations and a more informative contrastive learning approach for LiDAR segmentation.

Findings

Achieves state-of-the-art results on public benchmarks.

Improves consistency learning effectiveness.

Demonstrates significant performance gains over previous methods.

Abstract

The costly and time-consuming annotation process to produce large training sets for modelling semantic LiDAR segmentation methods has motivated the development of semi-supervised learning (SSL) methods. However, such SSL approaches often concentrate on employing consistency learning only for individual LiDAR representations. This narrow focus results in limited perturbations that generally fail to enable effective consistency learning. Additionally, these SSL approaches employ contrastive learning based on the sampling from a limited set of positive and negative embedding samples. This paper introduces a novel semi-supervised LiDAR semantic segmentation framework called ItTakesTwo (IT2). IT2 is designed to ensure consistent predictions from peer LiDAR representations, thereby improving the perturbation effectiveness in consistency learning. Furthermore, our contrastive learning employs informative samples drawn from a distribution of positive and negative embeddings learned from the entire training set. Results on public benchmarks show that our approach achieves remarkable improvements over the previous state-of-the-art (SOTA) methods in the field. The code is available at: https://github.com/yyliu01/IT2.