Continuous Facial Motion Deblurring

TL;DR

This paper presents a novel GAN-based framework for continuous facial motion deblurring from a single image, capable of restoring a sequence of sharp moments by controlling a moment factor, outperforming existing methods.

Contribution

It introduces a single-network, single-stage GAN framework with a moment control factor, facial motion reordering, and an auxiliary regressor for improved continuous face deblurring.

Findings

Generates various sharp frames by adjusting the control factor.

Outperforms recent single-to-video deblurring methods in perceptual metrics.

Achieves superior qualitative and quantitative results on 300VW dataset.

Abstract

We introduce a novel framework for continuous facial motion deblurring that restores the continuous sharp moment latent in a single motion-blurred face image via a moment control factor. Although a motion-blurred image is the accumulated signal of continuous sharp moments during the exposure time, most existing single image deblurring approaches aim to restore a fixed number of frames using multiple networks and training stages. To address this problem, we propose a continuous facial motion deblurring network based on GAN (CFMD-GAN), which is a novel framework for restoring the continuous moment latent in a single motion-blurred face image with a single network and a single training stage. To stabilize the network training, we train the generator to restore continuous moments in the order determined by our facial motion-based reordering process (FMR) utilizing domain-specific knowledge of the face. Moreover, we propose an auxiliary regressor that helps our generator produce more accurate images by estimating continuous sharp moments. Furthermore, we introduce a control-adaptive (ContAda) block that performs spatially deformable convolution and channel-wise attention as a function of the control factor. Extensive experiments on the 300VW datasets demonstrate that the proposed framework generates a various number of continuous output frames by varying the moment control factor. Compared with the recent single-to-single image deblurring networks trained with the same 300VW training set, the proposed method show the superior performance in restoring the central sharp frame in terms of perceptual metrics, including LPIPS, FID and Arcface identity distance. The proposed method outperforms the existing single-to-video deblurring method for both qualitative and quantitative comparisons.