Collaborative Propagation on Multiple Instance Graphs for 3D Instance Segmentation with Single-point Supervision

TL;DR

This paper introduces RWSeg, a weakly supervised 3D instance segmentation method that uses minimal labeling and propagates information through a novel cross-graph competing random walk algorithm, achieving competitive results.

Contribution

The paper presents a new weakly supervised framework for 3D instance segmentation using single-point labels and a cross-graph competing random walk algorithm for improved accuracy.

Findings

Achieves comparable performance to fully-supervised methods.

Outperforms previous weakly-supervised approaches.

Effective in propagating instance information with minimal labels.

Abstract

Instance segmentation on 3D point clouds has been attracting increasing attention due to its wide applications, especially in scene understanding areas. However, most existing methods operate on fully annotated data while manually preparing ground-truth labels at point-level is very cumbersome and labor-intensive. To address this issue, we propose a novel weakly supervised method RWSeg that only requires labeling one object with one point. With these sparse weak labels, we introduce a unified framework with two branches to propagate semantic and instance information respectively to unknown regions using self-attention and a cross-graph random walk method. Specifically, we propose a Cross-graph Competing Random Walks (CRW) algorithm that encourages competition among different instance graphs to resolve ambiguities in closely placed objects, improving instance assignment accuracy. RWSeg generates high-quality instance-level pseudo labels. Experimental results on ScanNet-v2 and S3DIS datasets show that our approach achieves comparable performance with fully-supervised methods and outperforms previous weakly-supervised methods by a substantial margin.