Rethinking Coarse-to-Fine Approach in Single Image Deblurring

TL;DR

This paper introduces MIMO-UNet, a novel single network architecture for image deblurring that efficiently combines multi-scale inputs and outputs with asymmetric feature fusion, outperforming existing methods in accuracy and speed.

Contribution

The paper proposes MIMO-UNet, a multi-input multi-output U-net that simplifies coarse-to-fine deblurring, reducing computational costs while improving performance.

Findings

Outperforms state-of-the-art methods on GoPro and RealBlur datasets.

Achieves higher accuracy with lower computational complexity.

Efficient multi-scale feature fusion enhances deblurring quality.

Abstract

Coarse-to-fine strategies have been extensively used for the architecture design of single image deblurring networks. Conventional methods typically stack sub-networks with multi-scale input images and gradually improve sharpness of images from the bottom sub-network to the top sub-network, yielding inevitably high computational costs. Toward a fast and accurate deblurring network design, we revisit the coarse-to-fine strategy and present a multi-input multi-output U-net (MIMO-UNet). The MIMO-UNet has three distinct features. First, the single encoder of the MIMO-UNet takes multi-scale input images to ease the difficulty of training. Second, the single decoder of the MIMO-UNet outputs multiple deblurred images with different scales to mimic multi-cascaded U-nets using a single U-shaped network. Last, asymmetric feature fusion is introduced to merge multi-scale features in an efficient manner. Extensive experiments on the GoPro and RealBlur datasets demonstrate that the proposed network outperforms the state-of-the-art methods in terms of both accuracy and computational complexity. Source code is available for research purposes at https://github.com/chosj95/MIMO-UNet.