Motion In-Betweening for Densely Interacting Characters

2025-09-30

AI Summary

This work addresses the long-horizon motion interpolation problem in dense two-person interactive scenarios. We propose a cross-space interpolation framework that generates natural and stable transitional motions while preserving precise spatiotemporal alignment of key poses. Methodologically, we jointly model conditional pose representations and periodic interaction patterns; incorporate adversarial learning to capture dynamic coupling regularities; and perform latent-space optimization to suppress pose drift accumulation, thereby enhancing long-term interaction consistency. Compared with existing approaches, our method significantly improves motion naturalness and interaction fidelity, enabling high-quality, controllable synthesis in complex scenarios such as boxing and dance. Quantitative evaluations and user studies demonstrate superior performance over state-of-the-art methods in motion quality, interaction coordination, and temporal stability.

Technology Category

• Humans and AI: Interaction Techniques and Devices
• Multiagent Systems: Adversarial Agents
• Intelligent Robots: Motion and Path Planning

Application Category

• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Motion in-betweening is the problem to synthesize movement between keyposes. Traditional research focused primarily on single characters. Extending them to densely interacting characters is highly challenging, as it demands precise spatial-temporal correspondence between the characters to maintain the interaction, while creating natural transitions towards predefined keyposes. In this research, we present a method for long-horizon interaction in-betweening that enables two characters to engage and respond to one another naturally. To effectively represent and synthesize interactions, we propose a novel solution called Cross-Space In-Betweening, which models the interactions of each character across different conditioning representation spaces. We further observe that the significantly increased constraints in interacting characters heavily limit the solution space, leading to degraded motion quality and diminished interaction over time. To enable long-horizon synthesis, we present two solutions to maintain long-term interaction and motion quality, thereby keeping synthesis in the stable region of the solution space.We first sustain interaction quality by identifying periodic interaction patterns through adversarial learning. We further maintain the motion quality by learning to refine the drifted latent space and prevent pose error accumulation. We demonstrate that our approach produces realistic, controllable, and long-horizon in-between motions of two characters with dynamic boxing and dancing actions across multiple keyposes, supported by extensive quantitative evaluations and user studies.

Problem

• Synthesizing motion between keyposes for multiple interacting characters
• Maintaining precise spatial-temporal correspondence during character interactions
• Enabling long-horizon natural transitions while preventing quality degradation

Innovation

• Cross-Space In-Betweening models interactions across representation spaces
• Adversarial learning identifies periodic patterns for sustained interaction
• Latent space refinement prevents pose error accumulation in motions