RASA: Disentangled Spatial-Motional Priors for Cross-Identity Character Animation
Computer Vision and Pattern Recognition
Summary
The authors present RASA, a method to animate a new character so it mimics the movements of another video while keeping correct size and position. They separate the problem into two parts: first, aligning the new character’s shape and position with the moving character using a Spatial Prior Calibrator, and second, controlling realistic joint movements through an Inherent Motional Guider. They test their method on a new benchmark called CIM-Bench and show it works better than previous techniques for realistic and well-aligned animations. This work highlights the importance of handling spatial alignment and motion separately for better character animation.
Authors
Zhen Xiao, Zhen Shen, Zhaofan Qiu, Ting Yao, Xueliang Liu, Tao Mei
Abstract
Cross-identity character animation aims to drive a target identity from a reference image to follow the motion of a source character from a driving video. The core challenge lies in the inherent entanglement of two capabilities: cross-identity spatial mapping (aligning position, scale, and skeletal proportions) and motion control (refining joint articulation, volumetric consistency, and view coherence). We introduce Reference-Aware Structural Alignment (RASA), a framework that disentangles spatial mapping from motion control by injecting structured priors into a Diffusion Transformer (DiT). Our approach has two stages. First, a Spatial Prior Calibrator (SPC) fuses reference identity with driving pose to generate a spatially grounded initial noise latent, ensuring correct positioning, scaling, and alignment with the driving skeleton. Second, an Inherent Motional Guider (IMG) encodes shape-agnostic SMPL articulation parameters into a semantic motion vector beyond appearance-biased 2D keypoints. Injected into intermediate DiT layers, this vector complements the base pose condition for anatomically consistent articulation and view-aware volumetric refinement. We curate CIM-Bench, a high-quality benchmark with rigorous curation, for evaluation. Extensive experiments show RASA significantly outperforms state-of-the-art methods in motion fidelity and visual quality. Our work establishes a new paradigm showing disentangled spatial and motional priors are key to robust character animation. Project page: https://hidream.ai.github.io/RASA/