GenHOI: Towards Object-Consistent Hand-Object Interaction with Temporally Balanced and Spatially Selective Object Injection
This work provides a significant improvement for researchers and developers working on realistic digital human video synthesis, particularly in maintaining object consistency during hand-object interactions, which is a known bottleneck.
This paper addresses the challenge of generating consistent hand-object interactions (HOI) in digital human videos, where existing methods struggle with object identity and physical plausibility in complex, in-the-wild scenarios. The authors propose GenHOI, an augmentation for pretrained video generation models, which significantly outperforms state-of-the-art HOI reenactment and all-in-one video editing methods in qualitative and quantitative evaluations on unseen, in-the-wild scenes.
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/