Attention based Broadly Self-guided Network for Low light Image Enhancement

TL;DR

This paper introduces ABSGN, an attention-based broadly self-guided network that efficiently enhances low-light images by effectively handling noise across various exposures, outperforming existing methods.

Contribution

The paper proposes a novel attention-based broadly self-guided network (ABSGN) that reduces inference time while improving feature extraction for low-light image enhancement.

Findings

Outperforms most state-of-the-art low-light enhancement methods

Effectively handles noise at different exposures

Reduces inference time compared to existing deep networks

Abstract

During the past years,deep convolutional neural networks have achieved impressive success in low-light Image Enhancement.Existing deep learning methods mostly enhance the ability of feature extraction by stacking network structures and deepening the depth of the network.which causes more runtime cost on single image.In order to reduce inference time while fully extracting local features and global features.Inspired by SGN,we propose a Attention based Broadly self-guided network (ABSGN) for real world low-light image Enhancement.such a broadly strategy is able to handle the noise at different exposures.The proposed network is validated by many mainstream benchmark.Additional experimental results show that the proposed network outperforms most of state-of-the-art low-light image Enhancement solutions.