SegGroup: Seg-Level Supervision for 3D Instance and Semantic Segmentation

TL;DR

SegGroup introduces a weakly-supervised approach for 3D scene segmentation that requires only one click per instance for annotation, significantly reducing labeling effort while maintaining high accuracy.

Contribution

The paper proposes a novel seg-level supervision method that leverages minimal annotations and a segment grouping network to generate pseudo labels for effective 3D segmentation.

Findings

Achieves comparable results to fully supervised methods.

Outperforms recent weakly-supervised approaches under the same annotation budget.

Reduces annotation cost significantly while maintaining accuracy.

Abstract

Most existing point cloud instance and semantic segmentation methods rely heavily on strong supervision signals, which require point-level labels for every point in the scene. However, such strong supervision suffers from large annotation costs, arousing the need to study efficient annotating. In this paper, we discover that the locations of instances matter for both instance and semantic 3D scene segmentation. By fully taking advantage of locations, we design a weakly-supervised point cloud segmentation method that only requires clicking on one point per instance to indicate its location for annotation. With over-segmentation for pre-processing, we extend these location annotations into segments as seg-level labels. We further design a segment grouping network (SegGroup) to generate point-level pseudo labels under seg-level labels by hierarchically grouping the unlabeled segments into the relevant nearby labeled segments, so that existing point-level supervised segmentation models can directly consume these pseudo labels for training. Experimental results show that our seg-level supervised method (SegGroup) achieves comparable results with the fully annotated point-level supervised methods. Moreover, it outperforms the recent weakly-supervised methods given a fixed annotation budget. Code is available at https://github.com/AnTao97/SegGroup.