Image deblurring based on lightweight multi-information fusion network

TL;DR

This paper introduces a lightweight multi-information fusion network (LMFN) for image deblurring that achieves state-of-the-art results with fewer parameters by combining multi-scale feature extraction and attention-based information fusion.

Contribution

The paper proposes a novel lightweight encoder-decoder network with multi-scale feature fusion and attention mechanisms for efficient image deblurring.

Findings

Achieves state-of-the-art deblurring performance

Uses fewer parameters than existing methods

Outperforms in model complexity and accuracy

Abstract

Recently, deep learning based image deblurring has been well developed. However, exploiting the detailed image features in a deep learning framework always requires a mass of parameters, which inevitably makes the network suffer from high computational burden. To solve this problem, we propose a lightweight multiinformation fusion network (LMFN) for image deblurring. The proposed LMFN is designed as an encoder-decoder architecture. In the encoding stage, the image feature is reduced to various smallscale spaces for multi-scale information extraction and fusion without a large amount of information loss. Then, a distillation network is used in the decoding stage, which allows the network benefit the most from residual learning while remaining sufficiently lightweight. Meanwhile, an information fusion strategy between distillation modules and feature channels is also carried out by attention mechanism. Through fusing different information in the proposed approach, our network can achieve state-of-the-art image deblurring result with smaller number of parameters and outperforms existing methods in model complexity.