Blur More To Deblur Better: Multi-Blur2Deblur For Efficient Video Deblurring

TL;DR

This paper introduces MB2D, a novel multi-blur approach for video deblurring that leverages multiple blurred frames and a recurrent neural network to improve deblurring performance efficiently.

Contribution

The paper proposes a new multi-blur-to-deblur framework with a multi-blurring recurrent neural network and multi-scale deblurring, enhancing efficiency and accuracy in video deblurring.

Findings

Achieved state-of-the-art results on GoPro and Su datasets.

Improved deblurring performance with multi-blurring recurrent neural network.

Enhanced efficiency in video deblurring methods.

Abstract

One of the key components for video deblurring is how to exploit neighboring frames. Recent state-of-the-art methods either used aligned adjacent frames to the center frame or propagated the information on past frames to the current frame recurrently. Here we propose multi-blur-to-deblur (MB2D), a novel concept to exploit neighboring frames for efficient video deblurring. Firstly, inspired by unsharp masking, we argue that using more blurred images with long exposures as additional inputs significantly improves performance. Secondly, we propose multi-blurring recurrent neural network (MBRNN) that can synthesize more blurred images from neighboring frames, yielding substantially improved performance with existing video deblurring methods. Lastly, we propose multi-scale deblurring with connecting recurrent feature map from MBRNN (MSDR) to achieve state-of-the-art performance on the popular GoPro and Su datasets in fast and memory efficient ways.