Reenact Anything: Semantic Video Motion Transfer Using Motion-Textual Inversion

TL;DR

This paper introduces motion-textual inversion, a novel method for semantic video motion transfer that uses motion embeddings with a pre-trained image-to-video model to generate realistic videos with controlled motion from a reference video.

Contribution

The paper proposes a new approach that leverages motion-textual embeddings and a pre-trained image-to-video model to improve semantic video motion transfer without requiring spatial alignment.

Findings

Outperforms existing methods in semantic video motion transfer

Enables high temporal motion granularity with multiple embedding tokens per frame

Generalizes across various domains and tasks such as reenactment and object motion control

Abstract

Recent years have seen a tremendous improvement in the quality of video generation and editing approaches. While several techniques focus on editing appearance, few address motion. Current approaches using text, trajectories, or bounding boxes are limited to simple motions, so we specify motions with a single motion reference video instead. We further propose to use a pre-trained image-to-video model rather than a text-to-video model. This approach allows us to preserve the exact appearance and position of a target object or scene and helps disentangle appearance from motion. Our method, called motion-textual inversion, leverages our observation that image-to-video models extract appearance mainly from the (latent) image input, while the text/image embedding injected via cross-attention predominantly controls motion. We thus represent motion using text/image embedding tokens. By operating on an inflated motion-text embedding containing multiple text/image embedding tokens per frame, we achieve a high temporal motion granularity. Once optimized on the motion reference video, this embedding can be applied to various target images to generate videos with semantically similar motions. Our approach does not require spatial alignment between the motion reference video and target image, generalizes across various domains, and can be applied to various tasks such as full-body and face reenactment, as well as controlling the motion of inanimate objects and the camera. We empirically demonstrate the effectiveness of our method in the semantic video motion transfer task, significantly outperforming existing methods in this context. Project website: https://mkansy.github.io/reenact-anything/