Eulerian Motion Guidance: Robust Image Animation via Bidirectional Geometric Consistency

2026-05-07

AI Summary

Existing diffusion-based image animation methods rely on Lagrangian motion guidance, which often suffers from low training efficiency, temporal inconsistency, and dynamic artifacts. This work proposes a local motion guidance strategy based on an Eulerian motion field, integrated with a bidirectional geometric consistency mechanism. By performing forward–backward cycle checks, the method identifies and masks occluded regions, effectively suppressing drift artifacts. The approach enables parallelized training and provides bounded-error supervision, significantly accelerating convergence while substantially improving temporal coherence and reducing dynamic artifacts.

Abstract

Recent advancements in image animation have utilized diffusion models to breathe life into static images. However, existing controllable frameworks typically rely on Lagrangian motion guidance, where optical flow is estimated relative to the initial frame. This paper revisits the same optical-flow primitive through a more local supervision design: we use adjacent-frame Eulerian motion fields to guide generation, where the motion signal always describes a short temporal hop. This shift enables parallelized training and provides bounded-error supervision throughout the generation process. To mitigate the drift artifacts common in adjacent frame generation, we introduce a Bidirectional Geometric Consistency mechanism, which computes a forward-backward cycle check to mathematically identify and mask occluded regions, preventing the model from learning incorrect warping objectives. Extensive experiments demonstrate that our approach accelerates training, preserves temporal coherence, and reduces dynamic artifacts compared to reference-based baselines.

Problem

• image animation
• motion guidance
• temporal coherence
• drift artifacts
• optical flow

Innovation

• Eulerian motion guidance
• bidirectional geometric consistency
• diffusion-based image animation
• optical flow supervision
• temporal coherence