UniFaceGAN: A Unified Framework for Temporally Consistent Facial Video Editing

TL;DR

UniFaceGAN is a unified framework that enables temporally consistent facial video editing, including face swapping and reenactment, by leveraging 3D reconstruction, a novel temporal loss, and region-aware normalization for more realistic results.

Contribution

The paper introduces a unified framework for facial video editing that handles multiple tasks simultaneously with improved temporal consistency and visual quality.

Findings

Produces more photo-realistic video portraits

Achieves smoother temporal consistency in edits

Outperforms state-of-the-art methods in quality

Abstract

Recent research has witnessed advances in facial image editing tasks including face swapping and face reenactment. However, these methods are confined to dealing with one specific task at a time. In addition, for video facial editing, previous methods either simply apply transformations frame by frame or utilize multiple frames in a concatenated or iterative fashion, which leads to noticeable visual flickers. In this paper, we propose a unified temporally consistent facial video editing framework termed UniFaceGAN. Based on a 3D reconstruction model and a simple yet efficient dynamic training sample selection mechanism, our framework is designed to handle face swapping and face reenactment simultaneously. To enforce the temporal consistency, a novel 3D temporal loss constraint is introduced based on the barycentric coordinate interpolation. Besides, we propose a region-aware conditional normalization layer to replace the traditional AdaIN or SPADE to synthesize more context-harmonious results. Compared with the state-of-the-art facial image editing methods, our framework generates video portraits that are more photo-realistic and temporally smooth.