# Texture-Aware Superpixel Segmentation

## TL;DR

This paper introduces TASP, a superpixel segmentation method that adaptively considers local texture information to improve segmentation accuracy in textured and smooth regions, outperforming existing methods.

## Contribution

The paper presents a novel Texture-Aware SuperPixel (TASP) algorithm that automatically adjusts spatial constraints based on local texture variance and uses patch-based distances for better texture homogeneity.

## Key findings

- TASP achieves higher segmentation accuracy on texture datasets.
- TASP outperforms state-of-the-art superpixel algorithms.
- TASP effectively balances spatial and texture features automatically.

## Abstract

Most superpixel algorithms compute a trade-off between spatial and color features at the pixel level. Hence, they may need fine parameter tuning to balance the two measures, and highly fail to group pixels with similar local texture properties. In this paper, we address these issues with a new Texture-Aware SuperPixel (TASP) method. To accurately segment textured and smooth areas, TASP automatically adjusts its spatial constraint according to the local feature variance. Then, to ensure texture homogeneity within superpixels, a new pixel to superpixel patch-based distance is proposed. TASP outperforms the segmentation accuracy of the state-of-the-art methods on texture and also natural color image datasets.