THOM: Generating Physically Plausible Hand-Object Meshes From Text
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866911586500739072 |
|---|---|
| author | Jeong, Uyoung Tiruneh, Yihalem Yimolal Chang, Hyung Jin Baek, Seungryul Kim, Kwang In |
| author_facet | Jeong, Uyoung Tiruneh, Yihalem Yimolal Chang, Hyung Jin Baek, Seungryul Kim, Kwang In |
| contents | Generating photorealistic 3D hand-object interactions (HOIs) from text is important for applications like robotic grasping and AR/VR content creation. In practice, however, achieving both visual fidelity and physical plausibility remains difficult, as mesh extraction from text-generated Gaussians is inherently ill-posed and the resulting meshes are often unreliable for physics-based optimization. We present THOM, a training-free framework that generates physically plausible 3D HOI meshes directly from text prompts, without requiring template object meshes. THOM follows a two-stage pipeline: it first generates hand and object Gaussians guided by text, and then refines their interaction using physics-based optimization. To enable reliable interaction modeling, we introduce a mesh extraction method with an explicit vertex-to-Gaussian mapping, which enables topology-aware regularization. We further improve physical plausibility through contact-aware optimization and vision-language model (VLM)-guided translation refinement. Extensive experiments show that THOM produces high-quality HOIs with strong text alignment, visual realism, and interaction plausibility. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_02736 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | THOM: Generating Physically Plausible Hand-Object Meshes From Text Jeong, Uyoung Tiruneh, Yihalem Yimolal Chang, Hyung Jin Baek, Seungryul Kim, Kwang In Computer Vision and Pattern Recognition Generating photorealistic 3D hand-object interactions (HOIs) from text is important for applications like robotic grasping and AR/VR content creation. In practice, however, achieving both visual fidelity and physical plausibility remains difficult, as mesh extraction from text-generated Gaussians is inherently ill-posed and the resulting meshes are often unreliable for physics-based optimization. We present THOM, a training-free framework that generates physically plausible 3D HOI meshes directly from text prompts, without requiring template object meshes. THOM follows a two-stage pipeline: it first generates hand and object Gaussians guided by text, and then refines their interaction using physics-based optimization. To enable reliable interaction modeling, we introduce a mesh extraction method with an explicit vertex-to-Gaussian mapping, which enables topology-aware regularization. We further improve physical plausibility through contact-aware optimization and vision-language model (VLM)-guided translation refinement. Extensive experiments show that THOM produces high-quality HOIs with strong text alignment, visual realism, and interaction plausibility. |
| title | THOM: Generating Physically Plausible Hand-Object Meshes From Text |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2604.02736 |