Diffusion Video Autoencoders: Toward Temporally Consistent Face Video Editing via Disentangled Video Encoding

TL;DR

This paper introduces a diffusion autoencoder framework for face video editing that ensures temporal consistency by disentangling identity and motion features, enabling robust and flexible editing of face videos.

Contribution

It is the first face video editing model to extract and manipulate disentangled identity and motion features using diffusion autoencoders for improved consistency and robustness.

Findings

Achieves temporally consistent face video editing.

Handles occlusions and wild face videos effectively.

Balances reconstruction and editing capabilities.

Abstract

Inspired by the impressive performance of recent face image editing methods, several studies have been naturally proposed to extend these methods to the face video editing task. One of the main challenges here is temporal consistency among edited frames, which is still unresolved. To this end, we propose a novel face video editing framework based on diffusion autoencoders that can successfully extract the decomposed features - for the first time as a face video editing model - of identity and motion from a given video. This modeling allows us to edit the video by simply manipulating the temporally invariant feature to the desired direction for the consistency. Another unique strength of our model is that, since our model is based on diffusion models, it can satisfy both reconstruction and edit capabilities at the same time, and is robust to corner cases in wild face videos (e.g. occluded faces) unlike the existing GAN-based methods.