UIS-Mamba: Exploring Mamba for Underwater Instance Segmentation via Dynamic Tree Scan and Hidden State Weaken

TL;DR

This paper introduces UIS-Mamba, a novel underwater instance segmentation model based on Mamba, featuring Dynamic Tree Scan and Hidden State Weaken modules to improve segmentation accuracy amidst underwater distortions.

Contribution

The paper presents the first Mamba-based model for underwater instance segmentation, with innovative modules to handle underwater scene challenges.

Findings

Achieves state-of-the-art results on UIIS and USIS10K datasets.

Maintains low parameters and computational complexity.

Effectively suppresses background interference.

Abstract

Underwater Instance Segmentation (UIS) tasks are crucial for underwater complex scene detection. Mamba, as an emerging state space model with inherently linear complexity and global receptive fields, is highly suitable for processing image segmentation tasks with long sequence features. However, due to the particularity of underwater scenes, there are many challenges in applying Mamba to UIS. The existing fixed-patch scanning mechanism cannot maintain the internal continuity of scanned instances in the presence of severely underwater color distortion and blurred instance boundaries, and the hidden state of the complex underwater background can also inhibit the understanding of instance objects. In this work, we propose the first Mamba-based underwater instance segmentation model UIS-Mamba, and design two innovative modules, Dynamic Tree Scan (DTS) and Hidden State Weaken (HSW), to migrate Mamba to the underwater task. DTS module maintains the continuity of the internal features of the instance objects by allowing the patches to dynamically offset and scale, thereby guiding the minimum spanning tree and providing dynamic local receptive fields. HSW module suppresses the interference of complex backgrounds and effectively focuses the information flow of state propagation to the instances themselves through the Ncut-based hidden state weakening mechanism. Experimental results show that UIS-Mamba achieves state-of-the-art performance on both UIIS and USIS10K datasets, while maintaining a low number of parameters and computational complexity. Code is available at https://github.com/Maricalce/UIS-Mamba.