Face Image Reflection Removal

TL;DR

This paper introduces a novel method for removing reflections from face images captured through glass, enhancing facial feature recovery and recognition accuracy by combining inpainting techniques with face-specific priors.

Contribution

It proposes a new reflection removal framework tailored for faces, utilizing a newly collected dataset and face priors to improve reflection removal and facial feature preservation.

Findings

Outperforms state-of-the-art reflection removal methods on face images.

Improves face recognition accuracy after reflection removal.

Demonstrates effectiveness on a newly collected face reflection dataset.

Abstract

Face images captured through the glass are usually contaminated by reflections. The non-transmitted reflections make the reflection removal more challenging than for general scenes, because important facial features are completely occluded. In this paper, we propose and solve the face image reflection removal problem. We remove non-transmitted reflections by incorporating inpainting ideas into a guided reflection removal framework and recover facial features by considering various face-specific priors. We use a newly collected face reflection image dataset to train our model and compare with state-of-the-art methods. The proposed method shows advantages in estimating reflection-free face images for improving face recognition.