First Order Motion Model for Image Animation

TL;DR

This paper introduces a self-supervised, category-agnostic image animation method that uses learned keypoints and local affine transformations to animate objects without prior annotations, achieving state-of-the-art results.

Contribution

It proposes a novel first order motion model that decouples appearance and motion, enabling flexible, annotation-free image animation across different object categories.

Findings

Outperforms existing methods on multiple benchmarks

Works effectively across diverse object categories

Handles complex motions with learned keypoints and affine transformations

Abstract

Image animation consists of generating a video sequence so that an object in a source image is animated according to the motion of a driving video. Our framework addresses this problem without using any annotation or prior information about the specific object to animate. Once trained on a set of videos depicting objects of the same category (e.g. faces, human bodies), our method can be applied to any object of this class. To achieve this, we decouple appearance and motion information using a self-supervised formulation. To support complex motions, we use a representation consisting of a set of learned keypoints along with their local affine transformations. A generator network models occlusions arising during target motions and combines the appearance extracted from the source image and the motion derived from the driving video. Our framework scores best on diverse benchmarks and on a variety of object categories. Our source code is publicly available.