Beyond Face Rotation: Global and Local Perception GAN for Photorealistic and Identity Preserving Frontal View Synthesis

TL;DR

This paper introduces TP-GAN, a novel two-pathway GAN that synthesizes photorealistic, identity-preserving frontal face images from profiles by combining global and local perception, improving face recognition accuracy.

Contribution

The paper proposes a new two-pathway GAN architecture with local patch networks and a combined loss function to better constrain the ill-posed frontal face synthesis problem.

Findings

Outperforms state-of-the-art face recognition methods on large pose datasets.

Produces more realistic and identity-preserving frontal face images.

Effectively combines global structure and local details for synthesis.

Abstract

Photorealistic frontal view synthesis from a single face image has a wide range of applications in the field of face recognition. Although data-driven deep learning methods have been proposed to address this problem by seeking solutions from ample face data, this problem is still challenging because it is intrinsically ill-posed. This paper proposes a Two-Pathway Generative Adversarial Network (TP-GAN) for photorealistic frontal view synthesis by simultaneously perceiving global structures and local details. Four landmark located patch networks are proposed to attend to local textures in addition to the commonly used global encoder-decoder network. Except for the novel architecture, we make this ill-posed problem well constrained by introducing a combination of adversarial loss, symmetry loss and identity preserving loss. The combined loss function leverages both frontal face distribution and pre-trained discriminative deep face models to guide an identity preserving inference of frontal views from profiles. Different from previous deep learning methods that mainly rely on intermediate features for recognition, our method directly leverages the synthesized identity preserving image for downstream tasks like face recognition and attribution estimation. Experimental results demonstrate that our method not only presents compelling perceptual results but also outperforms state-of-the-art results on large pose face recognition.