Welcome! I am Yihalem Yimolal Tiruneh, an M.S. researcher at the Vision & Learning Lab at the Ulsan National Institute of Science & Technology (UNIST), advised by Prof. Seungryul Baek and co-advised by Prof. Binod Bhattarai.

My research focuses on 3D human avatar reconstruction, 3D Gaussian splatting, and vision–language reasoning for hand and human understanding. My recent first-author work (ECCV 2026, under review) introduces a visibility-aware 3D Gaussian avatar framework that reconstructs animatable avatars from monocular video across full-body, upper-body, and head-only inputs within a single pipeline — combining occlusion-robust SMPL-X tracking, part-specific residual refinement, and diffusion-based texture completion.

I have additionally contributed to work on hand pose benchmarking (HandVQA, CVPR 2026), text-conditioned hand–object mesh generation (THOM, CVPR Findings 2026), real-time two-hand manipulation (QORT-Former, AAAI 2025), and Stable-Diffusion-based generative replay for class-incremental detection (SDDGR, CVPR 2024 Highlight, top 2.8%).

I earned my B.S. in Computer Science and Engineering from UNIST (Cum Laude, 2024), supported by the UNIST Dream Scholarship, and continue my M.S. on a Korean Government Graduate School Scholarship (full tuition + stipend).

News

• Apr 2026 — First-author paper Unified 3D Gaussian Human Avatars Under Arbitrary Body Visibility submitted to ECCV 2026.
• Feb 2026 — HandVQA accepted to CVPR 2026; THOM accepted to CVPR 2026 Findings.
• Dec 2024 — QORT-Former accepted to AAAI 2025.
• Feb 2024 — SDDGR accepted to CVPR 2024 as a Highlight (top 2.8%, 324 / 11,532 submissions).

Research interests

• 3D Gaussian Splatting & Neural Rendering
• 3D Human Avatar Reconstruction & Animation
• Visibility-Aware Optimization for Partial-View Inputs
• Diffusion Models for 3D Generation
• SMPL-X / FLAME Parametric Body Models
• Hand Pose Estimation & 3D Hand–Object Interaction
• Vision–Language Models for Spatial Reasoning

Get in touch

Feel free to reach out by email or connect on LinkedIn and GitHub. My full CV is available on the CV page and on Google Scholar.