Supervized Segmentation with Graph-Structured Deep Metric Learning

TL;DR

This paper introduces a supervised graph-structured deep metric learning approach for segmentation, utilizing a novel contrastive loss to produce embeddings that facilitate accurate graph partitioning, demonstrated on 3D point cloud oversegmentation.

Contribution

The paper proposes a new fully-supervised method with a graph-structured contrastive loss for learning vertex embeddings tailored for segmentation tasks.

Findings

Achieved state-of-the-art results on 3D point cloud oversegmentation

Introduced a novel contrastive loss structured by ground truth segmentation

Embeddings enable effective graph partitioning close to true segmentation

Abstract

We present a fully-supervized method for learning to segment data structured by an adjacency graph. We introduce the graph-structured contrastive loss, a loss function structured by a ground truth segmentation. It promotes learning vertex embeddings which are homogeneous within desired segments, and have high contrast at their interface. Thus, computing a piecewise-constant approximation of such embeddings produces a graph-partition close to the objective segmentation. This loss is fully backpropagable, which allows us to learn vertex embeddings with deep learning algorithms. We evaluate our methods on a 3D point cloud oversegmentation task, defining a new state-of-the-art by a large margin. These results are based on the published work of Landrieu and Boussaha 2019.