Image segmentation with superpixel-based covariance descriptors in low-rank representation

TL;DR

This paper presents two novel methods for image segmentation using superpixel covariance descriptors, employing Log-Euclidean metrics and low-rank representations to improve discriminative power, validated on the Berkeley dataset.

Contribution

The paper introduces two approaches that enhance superpixel covariance descriptors for segmentation by applying Log-Euclidean metrics and low-rank subspace modeling.

Findings

Both methods are competitive with state-of-the-art algorithms.

The Log-Euclidean kernel improves similarity measurement on covariance manifolds.

Low-rank representation captures subspace structures effectively.

Abstract

This paper investigates the problem of image segmentation using superpixels. We propose two approaches to enhance the discriminative ability of the superpixel's covariance descriptors. In the first one, we employ the Log-Euclidean distance as the metric on the covariance manifolds, and then use the RBF kernel to measure the similarities between covariance descriptors. The second method is focused on extracting the subspace structure of the set of covariance descriptors by extending a low rank representation algorithm on to the covariance manifolds. Experiments are carried out with the Berkly Segmentation Dataset, and compared with the state-of-the-art segmentation algorithms, both methods are competitive.