ECAFormer: Low-light Image Enhancement using Cross Attention

TL;DR

ECAFormer is a hierarchical transformer-based model for low-light image enhancement that effectively preserves details by enabling feature interaction across scales and layers, leading to improved performance.

Contribution

The paper introduces a novel hierarchical mutual enhancement architecture with cross-attention transformers and dual multi-head self-attention for better feature interaction in LLIE.

Findings

Nearly 3% PSNR improvement over existing methods

Effective preservation of image details across benchmarks

Demonstrates the importance of feature interaction in LLIE

Abstract

Low-light image enhancement (LLIE) is critical in computer vision. Existing LLIE methods often fail to discover the underlying relationships between different sub-components, causing the loss of complementary information between multiple modules and network layers, ultimately resulting in the loss of image details. To beat this shortage, we design a hierarchical mutual Enhancement via a Cross Attention transformer (ECAFormer), which introduces an architecture that enables concurrent propagation and interaction of multiple features. The model preserves detailed information by introducing a Dual Multi-head self-attention (DMSA), which leverages visual and semantic features across different scales, allowing them to guide and complement each other. Besides, a Cross-Scale DMSA block is introduced to capture the residual connection, integrating cross-layer information to further enhance image detail. Experimental results show that ECAFormer reaches competitive performance across multiple benchmarks, yielding nearly a 3% improvement in PSNR over the suboptimal method, demonstrating the effectiveness of information interaction in LLIE.