Single Source One Shot Reenactment using Weighted motion From Paired Feature Points
This addresses the problem of high-quality cross-person face animation for applications like video editing, though it appears incremental as it builds on existing reenactment methods.
The paper tackles the challenge of distinguishing facial motion from identity in cross-person face reenactment by proposing a model that learns shape-independent motion features using paired feature points, achieving faithful motion transfer while preserving source identity in extensive experiments.
Image reenactment is a task where the target object in the source image imitates the motion represented in the driving image. One of the most common reenactment tasks is face image animation. The major challenge in the current face reenactment approaches is to distinguish between facial motion and identity. For this reason, the previous models struggle to produce high-quality animations if the driving and source identities are different (cross-person reenactment). We propose a new (face) reenactment model that learns shape-independent motion features in a self-supervised setup. The motion is represented using a set of paired feature points extracted from the source and driving images simultaneously. The model is generalised to multiple reenactment tasks including faces and non-face objects using only a single source image. The extensive experiments show that the model faithfully transfers the driving motion to the source while retaining the source identity intact.