HOIGS: Human-Object Interaction Gaussian Splatting from Monocular Videos

ICLR 2026 Conference Submission17025 Authors

Keywords: Human object interaction reconstruction, 4D reconstruction

TL;DR: Human Object Interaction Scene Reconstruction

Abstract: Reconstructing dynamic scenes with complex human–object interactions is a fundamental challenge in computer vision and graphics. Existing Gaussian Splatting methods either rely on human pose priors, neglecting dynamic objects, or approximate all motions within a single field, limiting their ability to capture interaction-rich dynamics. To address this gap, we propose Human-Object Interaction Gaussian Splatting (HOIGS), which explicitly models interaction-induced deformation between humans and objects through a cross-attention based HOI module. Distinct deformation baselines are employed to extract complementary motion features: hexplane for humans and Cubic Hermite Spline (CHS) for objects. By integrating these heterogeneous features, HOIGS effectively captures interdependent motions and improves deformation estimation in scenarios involving occlusion, contact, and object manipulation. Comprehensive experiments on multiple datasets demonstrate that our method consistently outperforms state-of-the-art human-centric and 4D Gaussian approaches, highlighting the importance of explicitly modeling human–object interactions for high-fidelity reconstruction. The video results of HOIGS are available at: https://anonymous.4open.science/w/HOI-GS/

Primary Area: applications to computer vision, audio, language, and other modalities

Submission Number: 17025