Yuchen Lin | 林宇辰

Hi there! I am Yuchen Lin, a third-year undergraduate student at Peking University, majoring in Artificial Intelligence in Zhi Class (an honors program directed by Prof. Baoquan Chen) of EECS School. I am honored to be advised by Prof. Yadong Mu in Wangxuan Institute of Computer Technology, Peking University. I have spent a wonderful semester at School of Computer Science of Carnegie Mellon University as a visiting student. Since then, I have been working closely with Prof. Katerina Fragkiadaki in Machine Learning Department, Carnegie Mellon University.

I am actively looking for PhD opportunities starting in Fall 2026!

Email: linyuchen [at] stu.pku.edu.cn / yuchenlin0612 [at] gmail.com

GitHub / Google Scholar / Twitter / WeChat

News

[2025/07] One paper about 2D & 3D layout generation (InstructLayout) is accepted by TPAMI 2025.
[2025/01] One paper about 3D dynamics generation (OmniPhysGS) is accepted by ICLR 2025.
[2024/08] I come to CMU as a visiting student in School of Computer Science.

Research

My research interests lie in Computer Vision, epecially in the field of 3D asset understanding and generation.

*: equal contribution

	PartCrafter: Structured 3D Mesh Generation via Compositional Latent Diffusion Transformers Yuchen Lin, Chenguo Lin, Panwang Pan, Honglei Yan, Yiqiang Feng, Yadong Mu, Katerina Fragkiadaki arXiv, 2025 arxiv / code / website / youtube / PartCrafter is a structured 3D generative model that jointly generates multiple parts and objects from a single RGB image in one shot. PartCrafter received more than 1.5k stars within a month.
	Deformable Gaussian Diffusion: Controllable 4D Scene Generation from a Single Image Panwang Pan, Chenguo Lin, Jingjing Zhao, Chenxin Li, Yuchen Lin, Kairun Wen, Yunlong Lin, Yixuan Yuan, Yadong Mu, Zhiwen Fan arXiv, 2025 arxiv / code / website / Deformable Gaussian Diffusion is a generalizable framework for controllable 4D scene generation from a single image using a video diffusion model.
	MoVieS: Motion-Aware 4D Dynamic View Synthesis in One Second Chenguo Lin, Yuchen Lin, Panwang Pan, Yifan Yu, Honglei Yan, Katerina Fragkiadaki, Yadong Mu arXiv, 2025 arxiv / code / website / youtube / MoVieS is a feed-forward framework that jointly reconstructs appearance, geometry and motion for 4D scene perception from monocular videos.
	OmniPhysGS: 3D Constitutive Gaussians for General Physics-based Dynamics Generation Yuchen Lin, Chenguo Lin, Jianjin Xu, Yadong Mu International Conference on Learning Representations (ICLR), 2025 arxiv / code / website / OmniPhysGS is a novel framework for general physics-based 3D dynamic scene synthesis, which can automatically and flexibly model various materials with domain-expert Constitutive Gaussians.
	InstructLayout: Instruction-Driven 2D and 3D Layout Synthesis with Semantic Graph Prior Chenguo Lin, Yuchen Lin, Panwang Pan, Xuanyang Zhang, Yadong Mu Transactions on Pattern Analysis and Machine Intelligence (T-PAMI), 2025 arxiv / code / website / InstructLayout is a generative framework to both 2D E-commerce posters and 3D indoor scenes from instructions, and is composed of a semantic graph prior and a layout decoder.

Projects

These include coursework and other projects that I have worked on.

Weights2Weights++: Constructing the Weight Space for Customized Diffusion Models with VAE

Felipe Zanforlin Mautner, Yining She, Yuchen Lin
CMU 10423 Generative AI, 2024
This is the final project for 10423 during my exchange semester at CMU in 2024 Fall.
pdf / code / poster / group image /

Weights2Weights++ is an extention of Weights2Weights that constructs the weight space for customized diffusion models with VAE.

Experiences

	ByteDance PICO 2025.01 - Present Research Intern
	Carnegie Mellon University School of Computer Science 2024.08 - 2024.12 Exchange Student Machine Learning Department 2025.01 - Present Research Intern Research Advisor: Prof. Katerina Fragkiadaki
	Machine Intelligence Lab Wangxuan Institute of Computer Technology, Peking University 2023.09 - Present Research Intern Research Advisor: Prof. Yadong Mu
	Peking University School of Electronics Engineering and Computer Science 2022.09 - Present Undergraduate Student Academic Advisor: Prof. Baoquan Chen Major in Artificial Intelligence in Zhi Class, an Honors Program.

Services

I was the deputy director of the student union operations department in EECS School, Peking University.

I serve as a volunteer member of the Peking University admissions team every year.

Selected Awards

2024: Award for Scientific Research, Peking University

2024: Zhi Class Scholarship

2023: Robin Li Scholarship (25 awards per year in Peking University)

2023: Merit Student, Peking University

2022: XiaoSong Scholarship of Fuguang Foundation (4/200k+ every year in Fujian Province)

2020: First Prize, China Physics Olympiad (Fujian Province)

Miscellaneous

I love soccer. See how our team lifted the championship trophy 🏆 (a video with audio) in my high school, 2020, and achieved the runner-up 🥈 in the freshman cup of Peking University, 2022!

I am a big fan of Real Madrid. See my friends and I when we watched Madrid win the 2024 Champions League final. Many of them are fans of Barcelona! 😈

Design and source code from Jon Barron's website and Leonid Keselman's Jekyll fork.