GenHOI: Towards Object-Consistent Hand-Object Interaction with Temporally Balanced and Spatially Selective Object Injection

TL;DR

GenHOI introduces a novel augmentation technique for pretrained video models that enhances hand-object interaction realism and consistency in complex, in-the-wild scenarios by balancing temporal influence and spatial focus.

Contribution

It proposes Head-Sliding RoPE for temporal balancing and a two-level spatial attention gate for selective object injection, improving HOI generation in diverse scenes.

Findings

Outperforms state-of-the-art HOI reenactment methods

Enhances long-range object consistency in videos

Maintains background realism while improving interaction fidelity

Abstract

Hand-Object Interaction (HOI) remains a core challenge in digital human video synthesis, where models must generate physically plausible contact and preserve object identity across frames. Although recent HOI reenactment approaches have achieved progress, they are typically trained and evaluated in-domain and fail to generalize to complex, in-the-wild scenarios. In contrast, all-in-one video editing models exhibit broader robustness but still struggle with HOI-specific issues such as inconsistent object appearance. In this paper, we present GenHOI, a lightweight augmentation to pretrained video generation models that injects reference-object information in a temporally balanced and spatially selective manner. For temporal balancing, we propose Head-Sliding RoPE, which assigns head-specific temporal offsets to reference tokens, distributing their influence evenly across frames and mitigating the temporal decay of 3D RoPE to improve long-range object consistency. For spatial selectivity, we design a two-level spatial attention gate that concentrates object-conditioned attention on HOI regions and adaptively scales its strength, preserving background realism while enhancing interaction fidelity. Extensive qualitative and quantitative evaluations on unseen, in-the-wild scenes demonstrate that GenHOI significantly outperforms state-of-the-art HOI reenactment and all-in-one video editing methods. Project page: https://xuanhuang0.github.io/GenHOI/