MotionWeaver: Holistic 4D-Anchored Framework for Multi-Humanoid Image Animation

2026-02-11

AI Summary

Existing methods for animating character images struggle to generalize across diverse anthropomorphic characters due to challenges posed by morphological variability, complex interactions, and frequent occlusions. This work proposes an end-to-end framework that leverages an identity-agnostic unified motion representation and a holistic 4D anchoring paradigm to jointly embed motion and video latent representations within a shared 4D space. A hierarchical supervision mechanism is introduced to enhance modeling of interactions and occlusions. The approach achieves, for the first time, high-quality animation generation for a wide variety of anthropomorphic characters in complex multi-character scenarios. Evaluated on a newly curated benchmark of 300 videos, the method attains state-of-the-art performance, significantly outperforming existing approaches and demonstrating superior generalization capability.

Technology Category

• Humans and AI: Game Design — Virtual Humans, NPCs and Autonomous Characters
• Intelligent Robots: Manipulation
• Computer Vision: Motion & Tracking

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets

Abstract

Character image animation, which synthesizes videos of reference characters driven by pose sequences, has advanced rapidly but remains largely limited to single-human settings. Existing methods struggle to generalize to multi-humanoid scenarios, which involve diverse humanoid forms, complex interactions, and frequent occlusions. We address this gap with two key innovations. First, we introduce unified motion representations that extract identity-agnostic motions and explicitly bind them to corresponding characters, enabling generalization across diverse humanoid forms and seamless extension to multi-humanoid scenarios. Second, we propose a holistic 4D-anchored paradigm that constructs a shared 4D space to fuse motion representations with video latents, and further reinforces this process with hierarchical 4D-level supervision to better handle interactions and occlusions. We instantiate these ideas in MotionWeaver, an end-to-end framework for multi-humanoid image animation. To support this setting, we curate a 46-hour dataset of multi-human videos with rich interactions, and construct a 300-video benchmark featuring paired humanoid characters. Quantitative and qualitative experiments demonstrate that MotionWeaver not only achieves state-of-the-art results on our benchmark but also generalizes effectively across diverse humanoid forms, complex interactions, and challenging multi-humanoid scenarios.

Problem

• multi-humanoid animation
• pose-driven video synthesis
• occlusion handling
• character interaction
• motion generalization

Innovation

• unified motion representations
• 4D-anchored paradigm
• multi-humanoid animation
• hierarchical 4D supervision
• identity-agnostic motion