Animate Anyone 2: High-Fidelity Character Image Animation with Environment Affordance

TL;DR

Animate Anyone 2 advances character image animation by integrating environmental context, enabling more coherent and diverse animations through novel environmental representations, shape-agnostic masking, and object-guided feature blending.

Contribution

The paper introduces Animate Anyone 2, a novel method that incorporates environmental affordance into character animation, improving coherence and diversity over prior diffusion-based approaches.

Findings

Outperforms previous methods in animation quality.

Effectively models character-environment interactions.

Handles diverse motion patterns with pose modulation.

Abstract

Recent character image animation methods based on diffusion models, such as Animate Anyone, have made significant progress in generating consistent and generalizable character animations. However, these approaches fail to produce reasonable associations between characters and their environments. To address this limitation, we introduce Animate Anyone 2, aiming to animate characters with environment affordance. Beyond extracting motion signals from source video, we additionally capture environmental representations as conditional inputs. The environment is formulated as the region with the exclusion of characters and our model generates characters to populate these regions while maintaining coherence with the environmental context. We propose a shape-agnostic mask strategy that more effectively characterizes the relationship between character and environment. Furthermore, to enhance the fidelity of object interactions, we leverage an object guider to extract features of interacting objects and employ spatial blending for feature injection. We also introduce a pose modulation strategy that enables the model to handle more diverse motion patterns. Experimental results demonstrate the superior performance of the proposed method.