A Multi-Head Convolutional Neural Network With Multi-path Attention improves Image Denoising

TL;DR

This paper introduces MHCNN, a multi-head CNN with multi-path attention that leverages rotated images to enhance feature extraction and improve image denoising performance over existing models.

Contribution

The paper proposes a novel multi-head CNN with multi-path attention mechanism that utilizes rotated images to improve denoising, surpassing state-of-the-art methods.

Findings

MHCNN achieves higher PSNR than existing models.

The multi-path attention effectively integrates features from rotated images.

MHCNN performs well on both synthetic and real-world noise datasets.

Abstract

Recently, convolutional neural networks (CNNs) and attention mechanisms have been widely used in image denoising and achieved satisfactory performance. However, the previous works mostly use a single head to receive the noisy image, limiting the richness of extracted features. Therefore, a novel CNN with multiple heads (MH) named MHCNN is proposed in this paper, whose heads will receive the input images rotated by different rotation angles. MH makes MHCNN simultaneously utilize features of rotated images to remove noise. To integrate these features effectively, we present a novel multi-path attention mechanism (MPA). Unlike previous attention mechanisms that handle pixel-level, channel-level, or patch-level features, MPA focuses on features at the image level. Experiments show MHCNN surpasses other state-of-the-art CNN models on additive white Gaussian noise (AWGN) denoising and real-world image denoising. Its peak signal-to-noise ratio (PSNR) results are higher than other networks, such as BRDNet, RIDNet, PAN-Net, and CSANN. The code is accessible at https://github.com/JiaHongZ/MHCNN.