Improved Block Merging for 3D Point Cloud Instance Segmentation

TL;DR

This paper introduces a new block merging algorithm for 3D point cloud instance segmentation that corrects label errors and removes the need for instance overlap, leading to improved accuracy across various architectures.

Contribution

The work presents a novel block merging method that corrects label errors and eliminates the need for instance overlap, enhancing segmentation accuracy.

Findings

Significant accuracy improvements across all metrics

Effective correction of wrongly labelled points

Compatibility with various network architectures

Abstract

This paper proposes a novel block merging algorithm suitable for any block-based 3D instance segmentation technique. The proposed work improves over the state-of-the-art by allowing wrongly labelled points of already processed blocks to be corrected through label propagation. By doing so, instance overlap between blocks is not anymore necessary to produce the desirable results, which is the main limitation of the current art. Our experiments show that the proposed block merging algorithm significantly and consistently improves the obtained accuracy for all evaluation metrics employed in literature, regardless of the underlying network architecture.