HOIGS: Human-Object Interaction Gaussian Splatting
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866914447458566144 |
|---|---|
| author | Kim, Taewoo Yeom, Suwoong Pyun, Jaehyun Cha, Geonho Wee, Dongyoon Nam, Joonsik Jeong, Yun-Seong Kong, Kyeongbo Kang, Suk-Ju |
| author_facet | Kim, Taewoo Yeom, Suwoong Pyun, Jaehyun Cha, Geonho Wee, Dongyoon Nam, Joonsik Jeong, Yun-Seong Kong, Kyeongbo Kang, Suk-Ju |
| contents | 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 while 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 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_04016 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | HOIGS: Human-Object Interaction Gaussian Splatting Kim, Taewoo Yeom, Suwoong Pyun, Jaehyun Cha, Geonho Wee, Dongyoon Nam, Joonsik Jeong, Yun-Seong Kong, Kyeongbo Kang, Suk-Ju Computer Vision and Pattern Recognition Artificial Intelligence 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 while 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 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. |
| title | HOIGS: Human-Object Interaction Gaussian Splatting |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2604.04016 |