Efficient cascaded multiscale adaptive network for image restoration

• Main Authors: ZHOU, Yichen, ZHOU, Pan, NG, Teck Khim
• Format: text
• Language: English
• Published: Institutional Knowledge at Singapore Management University 2024
• Subjects: Image deblurring; Image denoising; Image super-resolution; Image restoration; Artificial Intelligence and Robotics; Numerical Analysis and Scientific Computing
• Online Access: https://ink.library.smu.edu.sg/sis_research/9641; https://ink.library.smu.edu.sg/context/sis_research/article/10641/viewcontent/Efficient_cascaded_multiscale_adaptive_network_for_image_restoration.pdf
• Institution: Singapore Management University

Image restoration, encompassing tasks such as deblurring, denoising, and super-resolution, remains a pivotal area in computer vision. However, efficiently addressing the spatially varying artifacts of various low-quality images with local adaptiveness and handling their degradations at different scales poses significant challenges. To efficiently tackle these issues, we propose the novel Efficient Cascaded Multiscale Adaptive (ECMA) Network. ECMA employs Local Adaptive Module, LAM, which dynamically adjusts convolution kernels across local image regions to efficiently handle varying artifacts. Thus, LAM addresses the local adaptiveness challenge more efficiently than costlier mechanisms like self-attention, due to its less computationally intensive convolutions. To construct a basic ECMA block, three cascading LAMs with convolution kernels from large to small sizes are employed to capture features at different scales. This cascaded multiscale learning effectively handles degradations at different scales, critical for diverse image restoration tasks. Finally, ECMA blocks are stacked in a U-Net architecture to build ECMA networks, which efficiently achieve both local adaptiveness and multiscale processing. Experiments show ECMA’s high performance and efficiency, achieving comparable or superior restoration performance to state-of-the-art methods while reducing computational costs by 1.2× to 9.7× across various image restoration tasks, e.g., image deblurring, denoising and super-resolution.