Splannequin: Freezing Monocular Mannequin-Challenge Footage with Dual-Detection Splatting
WACV 2025Splannequin freezes dynamic Gaussian splats into crisp 3D scenes from monocular videos by anchoring artifacts to more reliable temporal states.
Yi-Chuan Huang 黃怡川
Ph.D. Student · Department of Computer Science
National Yang Ming Chiao Tung University (NYCU), Taiwan
I am a third-year Ph.D. student in Computer Science at National Yang Ming Chiao Tung University (NYCU), advised by Prof. Yu-Lun Liu at the Computational Photography Lab. My research focused on 3D Computer Vision and Generative Models.
Outside of work and research, I enjoy cooking🧑🍳, caring for animals🐱, and working out🏃.
Research Interests
Computer Vision
3D Reconstruction
Neural Radiance Fields (NeRF)
3D Gaussian Splatting
Multi-view Diffusion Model
Novel View Synthesis
Generative AI
Object Generation & Editing
News
- Nov. 2025Paper accepted to WACV 2025: “Splannequin”. 🎉
- Jun. 2025Served as a reviewer for Pacific Graphics 2025 (PG 2025).
- Mar. 2025Paper accepted to CVPR 2025: “AuraFusion360”. 🎉
- Sep. 2024Awarded Outstanding Teaching Assistant for the course Signals and Systems. 📶
- Jun. 2024Passed Ph.D. Qualification and officially became a Ph.D. Candidate . 🧑🎓
- Sep. 2023Joined the Computational Photography Lab at NYCU as a Ph.D. student. 📖
Publications
AuraFusion360: Augmented Unseen Region Alignment for Reference-based 360° Unbounded Scene Inpainting
CVPR 2025Introduces depth-aware unseen mask generation, Adaptive Guided Depth Diffusion (zero-shot), and SDEdit-based detail enhancement for multi-view coherence.
Projects
Layered Vectorization of Natural Images for Editable SVG Graphics
Converted natural images into layered SVGs for intuitive and editable graphics. Achieved structure-preserving vectorization for AI-assisted design and editing.
Contact
Email: yichuanh.cs12@nycu.edu.tw