Eigencontours: Novel Contour Descriptors Based on Low-Rank Approximation

TL;DR

This paper introduces eigencontours, a new contour descriptor based on low-rank approximation, which effectively and efficiently represents object boundaries and improves instance segmentation performance.

Contribution

It proposes a novel low-rank based contour descriptor, eigencontours, and integrates it into an instance segmentation framework, showing superior boundary representation.

Findings

Eigencontours outperform existing descriptors in boundary representation.

The method achieves better segmentation accuracy on benchmark datasets.

Eigencontours enable low-dimensional, efficient boundary encoding.

Abstract

Novel contour descriptors, called eigencontours, based on low-rank approximation are proposed in this paper. First, we construct a contour matrix containing all object boundaries in a training set. Second, we decompose the contour matrix into eigencontours via the best rank-M approximation. Third, we represent an object boundary by a linear combination of the M eigencontours. We also incorporate the eigencontours into an instance segmentation framework. Experimental results demonstrate that the proposed eigencontours can represent object boundaries more effectively and more efficiently than existing descriptors in a low-dimensional space. Furthermore, the proposed algorithm yields meaningful performances on instance segmentation datasets.