Hierarchical Superpixel Segmentation via Structural Information Theory

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Xie, Minhui, Peng, Hao, Li, Pu, Zeng, Guangjie, Wang, Shuhai, Wu, Jia, Li, Peng, Yu, Philip S.
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866909455009972224
author Xie, Minhui
Peng, Hao
Li, Pu
Zeng, Guangjie
Wang, Shuhai
Wu, Jia
Li, Peng
Yu, Philip S.
author_facet Xie, Minhui
Peng, Hao
Li, Pu
Zeng, Guangjie
Wang, Shuhai
Wu, Jia
Li, Peng
Yu, Philip S.
contents Superpixel segmentation is a foundation for many higher-level computer vision tasks, such as image segmentation, object recognition, and scene understanding. Existing graph-based superpixel segmentation methods typically concentrate on the relationships between a given pixel and its directly adjacent pixels while overlooking the influence of non-adjacent pixels. These approaches do not fully leverage the global information in the graph, leading to suboptimal segmentation quality. To address this limitation, we present SIT-HSS, a hierarchical superpixel segmentation method based on structural information theory. Specifically, we first design a novel graph construction strategy that incrementally explores the pixel neighborhood to add edges based on 1-dimensional structural entropy (1D SE). This strategy maximizes the retention of graph information while avoiding an overly complex graph structure. Then, we design a new 2D SE-guided hierarchical graph partitioning method, which iteratively merges pixel clusters layer by layer to reduce the graph's 2D SE until a predefined segmentation scale is achieved. Experimental results on three benchmark datasets demonstrate that the SIT-HSS performs better than state-of-the-art unsupervised superpixel segmentation algorithms. The source code is available at \url{https://github.com/SELGroup/SIT-HSS}.
format Preprint
id arxiv_https___arxiv_org_abs_2501_07069
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Hierarchical Superpixel Segmentation via Structural Information Theory
Xie, Minhui
Peng, Hao
Li, Pu
Zeng, Guangjie
Wang, Shuhai
Wu, Jia
Li, Peng
Yu, Philip S.
Computer Vision and Pattern Recognition
Superpixel segmentation is a foundation for many higher-level computer vision tasks, such as image segmentation, object recognition, and scene understanding. Existing graph-based superpixel segmentation methods typically concentrate on the relationships between a given pixel and its directly adjacent pixels while overlooking the influence of non-adjacent pixels. These approaches do not fully leverage the global information in the graph, leading to suboptimal segmentation quality. To address this limitation, we present SIT-HSS, a hierarchical superpixel segmentation method based on structural information theory. Specifically, we first design a novel graph construction strategy that incrementally explores the pixel neighborhood to add edges based on 1-dimensional structural entropy (1D SE). This strategy maximizes the retention of graph information while avoiding an overly complex graph structure. Then, we design a new 2D SE-guided hierarchical graph partitioning method, which iteratively merges pixel clusters layer by layer to reduce the graph's 2D SE until a predefined segmentation scale is achieved. Experimental results on three benchmark datasets demonstrate that the SIT-HSS performs better than state-of-the-art unsupervised superpixel segmentation algorithms. The source code is available at \url{https://github.com/SELGroup/SIT-HSS}.
title Hierarchical Superpixel Segmentation via Structural Information Theory
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2501.07069