SSMamba: Superpixel Segmentation With Mamba
- 9 April 2025
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Signal Processing Letters
- Vol. 32 (10709908) , 1715-1719
Abstract
Deep convolutional networks have achieved remarkable success in superpixel segmentation. However, they only focus on local features ignoring global attributes. The visual Mamba demonstrates an exceptional capability to capture long-range dependencies and offers a lower computational cost compared to the Transformer. Building on this inspiration, we propose a novel superpixel segmentation with Mamba, termed SSMamba. In SSMamba, Mamba is integrated into a global-local architecture, enabling efficient interaction between global attributes and local features to produce high-quality superpixels. The designed activation function further enhances the effectiveness of SSMamba. Extensive experiments on four public datasets demonstrate that SSMamba outperforms existing state-of-the-art methods, achieving competitive average values of ASA = 0.9541, BR = 0.8768, BP = 0.2124, UE = 0.0910, and CO = 0.3698.Keywords
Funding Information
- National Natural Science Foundation of China (62366029)
- Gansu Province Youth Science and Technology Fund (23JRRA855)
- Key Laboratory of Big Data & Artificial Intelligence in Transportation
- Ministry of Education (BATLAB202302)
- Key Research and Development Project of Lanzhou Jiaotong University (ZDYF2304)
- Young Scholars Science Foundation of Lanzhou Jiaotong University (2023006)
This publication has 27 references indexed in Scilit:
- Superpixel-wise contrast exploration for salient object detectionKnowledge-Based Systems, 2024
- Superpixel Segmentation With Fully Convolutional Networks2020
- Sigmoid-weighted linear units for neural network function approximation in reinforcement learningNeural Networks, 2018
- Superpixel Sampling Networks2018
- Learning Superpixels with Segmentation-Aware Affinity Loss2018
- Augmented Reality Meets Computer Vision: Efficient Data Generation for Urban Driving ScenesInternational Journal of Computer Vision, 2018
- Superpixels and Polygons Using Simple Non-iterative Clustering2017
- U-Net: Convolutional Networks for Biomedical Image Segmentation2015
- Indoor Segmentation and Support Inference from RGBD Images2012
- Entropy rate superpixel segmentation2011