Demo of Face2Face

TLDR

The authors aim to animate a target video’s facial expressions using a source actor and render the result photorealistically. They achieve this by recovering facial identity from monocular video with non‑rigid bundling, tracking expressions via dense photometric consistency, transferring deformations, warping mouth interiors, and re‑rendering the synthesized face onto the live stream. The system is shown to reenact YouTube videos in real time during a live demonstration.

Abstract

We present a novel approach for real-time facial reenactment of a monocular target video sequence (e.g., Youtube video). The source sequence is also a monocular video stream, captured live with a commodity webcam. Our goal is to animate the facial expressions of the target video by a source actor and re-render the manipulated output video in a photo-realistic fashion. To this end, we first address the under-constrained problem of facial identity recovery from monocular video by non-rigid model-based bundling. At run time, we track facial expressions of both source and target video using a dense photometric consistency measure. Reenactment is then achieved by fast and efficient deformation transfer between source and target. The mouth interior that best matches the re-targeted expression is retrieved from the target sequence and warped to produce an accurate fit. Finally, we convincingly re-render the synthesized target face on top of the corresponding video stream such that it seamlessly blends with the real-world illumination. We demonstrate our method in a live setup, where Youtube videos are reen-acted in real time.