Unleashing Guidance Without Classifiers for Human-Object Interaction Animation

TL;DR

This paper introduces LIGHT, a novel diffusion-based method for human-object interaction animation that eliminates the need for handcrafted contact priors by using denoising pace as guidance, leading to more realistic and contact-aware animations.

Contribution

LIGHT presents a data-driven guidance approach in diffusion models for HOI animation, reducing reliance on manual priors and improving contact fidelity and generalization.

Findings

Outperforms classifier-free guidance in contact fidelity.

Achieves more realistic and diverse HOI animations.

Generalizes well to unseen objects and tasks.

Abstract

Generating realistic human-object interaction (HOI) animations remains challenging because it requires jointly modeling dynamic human actions and diverse object geometries. Prior diffusion-based approaches often rely on hand-crafted contact priors or human-imposed kinematic constraints to improve contact quality. We propose LIGHT, a data-driven alternative in which guidance emerges from the denoising pace itself, reducing dependence on manually designed priors. Building on diffusion forcing, we factor the representation into modality-specific components and assign individualized noise levels with asynchronous denoising schedules. In this paradigm, cleaner components guide noisier ones through cross-attention, yielding guidance without auxiliary classifiers. We find that this data-driven guidance is inherently contact-aware, and can be enhanced when training is augmented with a broad spectrum of synthetic object geometries, encouraging invariance of contact semantics to geometric diversity. Extensive experiments show that pace-induced guidance more effectively mirrors the benefits of contact priors than conventional classifier-free guidance, while achieving higher contact fidelity, more realistic HOI generation, and stronger generalization to unseen objects and tasks.