Yihalem Yimolal Tiruneh

Yihalem Yimolal Tiruneh

M.S. Researcher · Vision & Learning Lab, UNIST. 3D human avatar reconstruction, 3D Gaussian splatting, and vision–language models.

THOM: Generating Physically Plausible Hand-Object Meshes From Text

Published in CVPR 2026 Findings, 2026

Authors. U. Jeong, Y. Y. Tiruneh, H. J. Chang, S. Baek, K. I. Kim.

Venue. CVPR 2026 Findings.

Project page. uyoung-jeong.github.io/THOM_Project

Summary.

  • A text-conditioned generative framework that produces physically plausible hand–object mesh interactions from natural-language prompts.
  • Bridges language conditioning with explicit physical-plausibility constraints for downstream interaction synthesis and manipulation tasks.