CoopDiff: Anticipating 3D Human-object Interactions via Contact-consistent Decoupled Diffusion

2025-08-09

AI Summary

Existing methods commonly model human and object motion jointly, neglecting their distinct physical dynamics, thereby limiting the accuracy and physical plausibility of 3D human–object interaction (HOI) prediction. To address this, we propose CoopDiff, a decoupled diffusion framework. It employs a dual-branch architecture to separately model human pose and object trajectory, explicitly capturing their heterogeneous dynamics. We introduce, for the first time, a contact-point consistency constraint to enforce spatiotemporal coherence at interaction points. Additionally, we design a human-driven interaction guidance module to enhance cross-branch coordination. Through a joint denoising process, CoopDiff generates physically plausible HOI sequences that satisfy kinematic and contact constraints. Evaluated on BEHAVE and HOI benchmarks, CoopDiff achieves significant improvements over state-of-the-art methods in motion accuracy, contact stability, and physical realism.

Technology Category

• Humans and AI: Human-Aware Planning and Behavior Prediction
• Computer Vision: Diffusion Models for Vision
• Cognitive Modeling & Cognitive Systems: Simulating Human Behavior

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems

Abstract

3D human-object interaction (HOI) anticipation aims to predict the future motion of humans and their manipulated objects, conditioned on the historical context. Generally, the articulated humans and rigid objects exhibit different motion patterns, due to their distinct intrinsic physical properties. However, this distinction is ignored by most of the existing works, which intend to capture the dynamics of both humans and objects within a single prediction model. In this work, we propose a novel contact-consistent decoupled diffusion framework CoopDiff, which employs two distinct branches to decouple human and object motion modeling, with the human-object contact points as shared anchors to bridge the motion generation across branches. The human dynamics branch is aimed to predict highly structured human motion, while the object dynamics branch focuses on the object motion with rigid translations and rotations. These two branches are bridged by a series of shared contact points with consistency constraint for coherent human-object motion prediction. To further enhance human-object consistency and prediction reliability, we propose a human-driven interaction module to guide object motion modeling. Extensive experiments on the BEHAVE and Human-object Interaction datasets demonstrate that our CoopDiff outperforms state-of-the-art methods.

Problem

• Predict future 3D human-object interactions from historical context
• Decouple human and object motion modeling using contact points
• Enhance interaction consistency via human-driven object motion guidance

Innovation

• Decoupled diffusion for human-object motion modeling
• Contact points bridge motion generation branches
• Human-driven interaction module enhances consistency