PPMamba: A Pyramid Pooling Local Auxiliary SSM-Based Model for Remote Sensing Image Semantic Segmentation

TL;DR

PPMamba is a novel hybrid model combining pyramid pooling, CNN, and Mamba state space models to improve remote sensing image segmentation by capturing long-range dependencies and local features efficiently.

Contribution

It introduces the PP-SSM block that integrates local auxiliary mechanisms with omnidirectional state space modeling for enhanced segmentation.

Findings

Achieves competitive results on ISPRS Vaihingen dataset.

Outperforms several state-of-the-art models in accuracy.

Effectively captures long-range dependencies and local features.

Abstract

Semantic segmentation is a vital task in the field of remote sensing (RS). However, conventional convolutional neural network (CNN) and transformer-based models face limitations in capturing long-range dependencies or are often computationally intensive. Recently, an advanced state space model (SSM), namely Mamba, was introduced, offering linear computational complexity while effectively establishing long-distance dependencies. Despite their advantages, Mamba-based methods encounter challenges in preserving local semantic information. To cope with these challenges, this paper proposes a novel network called Pyramid Pooling Mamba (PPMamba), which integrates CNN and Mamba for RS semantic segmentation tasks. The core structure of PPMamba, the Pyramid Pooling-State Space Model (PP-SSM) block, combines a local auxiliary mechanism with an omnidirectional state space model (OSS) that selectively scans feature maps from eight directions, capturing comprehensive feature information. Additionally, the auxiliary mechanism includes pyramid-shaped convolutional branches designed to extract features at multiple scales. Extensive experiments on two widely-used datasets, ISPRS Vaihingen and LoveDA Urban, demonstrate that PPMamba achieves competitive performance compared to state-of-the-art models.