my photo

Yumeng He 何雨濛

PhD Student
University of California, Los Angeles
heyumeng0928@gmail.com
LinkedIn
Github
Google Scholar
Resume
WeChat

Last Updated: Aug 7, 2026

Total number of visits

About Me

I am a 1st year PhD student at the UCLA AIVC Lab, co-supervised by Prof. Chenfanfu Jiang and Prof. Demetri Terzopoulos. Previously, I obtained my Master's degree in Computer Science from the University of Southern California, supervised by Prof. Jernej Barbič, and my Honours Bachelor of Science degree with High Distinction from the University of Toronto, where I double-majored in Mathematics and Computer Science.

Research Interests

My research interests are computer vision, graphics, and robotics, with a focus on physical AI, object/scene generation, articulated and kinematic asset modeling, physics-based simulation, real-to-sim pipelines, and policy learning.
I am always open to collaborations and summer internship opportunities, feel free to drop me an email if you are interested in my research.
We are recruiting interns (remote or in-person), please reach out with your CV if you are interested!

News

Experience

ENVORA-logo
Envora Inc. | Research Intern Aug 2026 - Present
UCLA-logo
UCLA AIVC Lab | Visiting Student Jun 2025 - Jul 2026
Supervisor: Prof. Chenfanfu Jiang, Dr. Ying Jiang
RESL-logo
USC RESL Lab | Research Assistant Jul 2025 - Dec 2025
Supervisor: Prof. Gaurav S. Sukhatme, Dr. Jeremy Morgan
CDUT-logo
CDUT CIST | Research Assistant May 2023 - Aug 2023
Supervisor: Prof. Guangzhu Chen
HCL-logo
HCL Canada Inc. | Software Developer Intern Aug 2022 - Aug 2023

Publications & Manuscripts

Fishbone
Fishbone: From One 3D Asset to a Million Controllable Edits
Yumeng He, Xiaoying Wang, Peihao Li, Yanjia Huang, Joe Masterjohn, Jiajun Wu, Leonidas Guibas, Yin Yang, Ying Jiang, Chenfanfu Jiang
arXiv, 2026
[Project Page] [arXiv]
SeeClear
SeeClear: Reliable Transparent Object Depth Estimation via Generative Opacification
Xiaoying Wang*, Yumeng He*, Jingkai Shi*, Jiayin Lu, Yin Yang, Ying Jiang, Chenfanfu Jiang
European Conference on Computer Vision (ECCV), 2026
[Project Page] [arXiv]
VisHintPrompt
VisHintPrompt: A Visual Hint Prompting Strategy for Numerical Inference from Charts
Fengling Zheng, Yongle Peng, Zi Rong, Chenyun Cai, Yumeng He, Dekun Qian, Zhiguang Zhou, Yigang Wang, Yong Wang
2025
SPARK
SPARK: Sim-ready Part-level Articulated Reconstruction with VLM Knowledge
Yumeng He*, Ying Jiang*, Jiayin Lu*, Yin Yang, Chenfanfu Jiang
Computer Vision and Pattern Recognition (CVPR), 2026
Oral (top 3.45%) CVPR Compute Transparency Champion
[Project Page] [arXiv] [Code]
vorolight
VoroLight: Learning Voronoi Surface Meshes via Sphere Intersection
Jiayin Lu*, Ying Jiang*, Yumeng He*, Yin Yang, Chenfanfu Jiang
arXiv, 2025
[Project Page] [arXiv]
painting
Birth of a Painting: Differentiable Brushstroke Reconstruction
Ying Jiang*, Jiayin Lu*, Yunuo Chen*, Yumeng He, Kui Wu, Yin Yang, Chenfanfu Jiang
Conditionally Accepted by ACM Transactions On Graphics (SIGGRAPH), 2026
[Project Page] [arXiv]
CDUT
Self-supervised dual-layer 2D normalizing flow method for industrial anomaly detection
Zhenlian Miao, Guangzhu Chen, Xiaojuan Liao, Jiu Dai, Yumeng He
Applied Soft Computing (ASOC), 2024
[Paper] [Code]

Master's Thesis

VCShell
From Kirchhoff-Love Shells to Volumetrically Consistent Shells: Understanding the Limitations of Classical Thin-Shell Theory and Developing the Two-Mode VC Shells
Yumeng He, Jernej Barbic
Master's Thesis, 2026
Committee: Jernej Barbic, Chenfanfu Jiang, Carter Slocum
[Paper]

Service & Activities

USC Courses

Fun Facts