Dynamic Spectral Residual Superpixels

Abstract

We consider the problem of segmenting an image into superpixels in the context of $k$-means clustering, in which we wish to decompose an image into local, homogeneous regions corresponding to the underlying objects. Our novel approach builds upon the widely used Simple Linear Iterative Clustering (SLIC), and incorporate a measure of objects' structure based on the spectral residual of an image. Based on this combination, we propose a modified initialisation scheme and search metric, which helps keeps fine-details. This combination leads to better adherence to object boundaries, while preventing unnecessary segmentation of large, uniform areas, while remaining computationally tractable in comparison to other methods. We demonstrate through numerical and visual experiments that our approach outperforms the state-of-the-art techniques.

References in corpus (5)

• Superpixels: An Evaluation of the State-of-the-Art
• Recent progress in semantic image segmentation
• Accurate Light Field Depth Estimation with Superpixel Regularization over Partially Occluded Regions
• Superpixel Contracted Graph-Based Learning for Hyperspectral Image Classification
• The Mutex Watershed and its Objective: Efficient, Parameter-Free Graph Partitioning

Cited by in corpus (1)

• A comprehensive review and new taxonomy on superpixel segmentation