Joint Face Image Restoration and Frontalization for Recognition
Xiaoguang Tu, Jian Zhao, Qiankun Liu, Wenjie Ai, Guodong Guo, Zhifeng Li, Wei Liu, Jiashi Feng
Abstract
In real-world scenarios, many factors may harm face recognition performance, <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">e.g</i> ., large pose, bad illumination, low resolution, blur and noise. To address these challenges, previous efforts usually first restore the low-quality faces to high-quality ones and then perform face recognition. However, most of these methods are stage-wise, which is sub-optimal and deviates from the reality. In this paper, we address all these challenges jointly for unconstrained face recognition. We propose an <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">M</b> ulti- <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">D</b> egradation <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">F</b> ace <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">R</b> estoration (MDFR) model to restore frontalized high-quality faces from the given low-quality ones under arbitrary facial poses, with three distinct novelties. First, MDFR is a well-designed encoder-decoder architecture which extracts feature representation from an input face image with arbitrary low-quality factors and restores it to a high-quality counterpart. Second, MDFR introduces a pose residual learning strategy along with a 3D-based <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">P</b> ose <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">N</b> ormalization <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">M</b> odule (PNM), which can perceive the pose gap between the input initial pose and its real-frontal pose to guide the face frontalization. Finally, MDFR can generate frontalized high-quality face images by a single unified network, showing a strong capability of preserving face identity. Qualitative and quantitative experiments on both controlled and in-the-wild benchmarks demonstrate the superiority of MDFR over state-of-the-art methods on both face frontalization and face restoration.