Texture-Aware Superpixel Segmentation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Giraud, Remi, Ta, Vinh-Thong, Papadakis, Nicolas, Berthoumieu, Yannick
Format: Preprint
Published: 2019
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912590221803520
author Giraud, Remi
Ta, Vinh-Thong
Papadakis, Nicolas
Berthoumieu, Yannick
author_facet Giraud, Remi
Ta, Vinh-Thong
Papadakis, Nicolas
Berthoumieu, Yannick
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_1901_11111
institution arXiv
publishDate 2019
record_format arxiv
spellingShingle Texture-Aware Superpixel Segmentation
Giraud, Remi
Ta, Vinh-Thong
Papadakis, Nicolas
Berthoumieu, Yannick
Computer Vision and Pattern Recognition
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.
title Texture-Aware Superpixel Segmentation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/1901.11111