EAMamba: Efficient All-Around Vision State Space Model for Image Restoration

2025-06-27

AI Summary

Vision Mamba shows promise for image restoration but suffers from quadratic computational complexity with sequence length and loss of local pixel information due to unidirectional scanning. To address these limitations, we propose EAMamba—an efficient omnidirectional visual state-space model. It introduces a multi-head selective scanning module to enable parallel multi-directional sequence modeling and an omnidirectional scanning mechanism that captures global dependencies in linear complexity while mitigating local forgetting. Crucially, EAMamba integrates multi-scale contextual information without increasing parameters or FLOPs. Evaluated on super-resolution, denoising, deblurring, and dehazing tasks, EAMamba reduces computational cost by 31%–89% in FLOPs while maintaining state-of-the-art restoration performance. This work significantly advances the efficiency and representational capacity of state-space models in low-level vision.

Technology Category

• Computer Vision: Multi-modal Vision
• Machine Learning: Large Multimodal Models (LMMs)
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems

Abstract

Image restoration is a key task in low-level computer vision that aims to reconstruct high-quality images from degraded inputs. The emergence of Vision Mamba, which draws inspiration from the advanced state space model Mamba, marks a significant advancement in this field. Vision Mamba demonstrates excellence in modeling long-range dependencies with linear complexity, a crucial advantage for image restoration tasks. Despite its strengths, Vision Mamba encounters challenges in low-level vision tasks, including computational complexity that scales with the number of scanning sequences and local pixel forgetting. To address these limitations, this study introduces Efficient All-Around Mamba (EAMamba), an enhanced framework that incorporates a Multi-Head Selective Scan Module (MHSSM) with an all-around scanning mechanism. MHSSM efficiently aggregates multiple scanning sequences, which avoids increases in computational complexity and parameter count. The all-around scanning strategy implements multiple patterns to capture holistic information and resolves the local pixel forgetting issue. Our experimental evaluations validate these innovations across several restoration tasks, including super resolution, denoising, deblurring, and dehazing. The results validate that EAMamba achieves a significant 31-89% reduction in FLOPs while maintaining favorable performance compared to existing low-level Vision Mamba methods.

Problem

• Addresses computational complexity in Vision Mamba for image restoration
• Solves local pixel forgetting issue in low-level vision tasks
• Reduces FLOPs significantly while maintaining performance

Innovation

• Multi-Head Selective Scan Module (MHSSM)
• All-around scanning mechanism
• Linear complexity long-range dependency modeling