Multi-dimension Queried and Interacting Network for Stereo Image Deraining

TL;DR

This paper introduces MQINet, a novel stereo image deraining model that uses multi-dimension queries and interactions, incorporating physics-aware attention to effectively reduce rain artifacts and outperform existing methods.

Contribution

The paper proposes a multi-dimension queried and interacting network with novel attention modules for improved stereo image deraining, integrating physics-based and cross-view feature interactions.

Findings

Achieves 4.18 dB higher PSNR than EPRRNet

Outperforms StereoIRR by 0.45 dB in PSNR

Demonstrates superior rain removal performance in stereo images

Abstract

Eliminating the rain degradation in stereo images poses a formidable challenge, which necessitates the efficient exploitation of mutual information present between the dual views. To this end, we devise MQINet, which employs multi-dimension queries and interactions for stereo image deraining. More specifically, our approach incorporates a context-aware dimension-wise queried block (CDQB). This module leverages dimension-wise queries that are independent of the input features and employs global context-aware attention (GCA) to capture essential features while avoiding the entanglement of redundant or irrelevant information. Meanwhile, we introduce an intra-view physics-aware attention (IPA) based on the inverse physical model of rainy images. IPA extracts shallow features that are sensitive to the physics of rain degradation, facilitating the reduction of rain-related artifacts during the early learning period. Furthermore, we integrate a cross-view multi-dimension interacting attention mechanism (CMIA) to foster comprehensive feature interaction between the two views across multiple dimensions. Extensive experimental evaluations demonstrate the superiority of our model over EPRRNet and StereoIRR, achieving respective improvements of 4.18 dB and 0.45 dB in PSNR. Code and models are available at \url{https://github.com/chdwyb/MQINet}.