Yuchen Lin | 林宇辰

ID-Crafter: VLM-Grounded Online RL for Compositional Multi-Subject Video Generation

Panwang Pan*, Jingjing Zhao*, Yuchen Lin*, Chenguo Lin, Chenxin Li, HengYu Liu, Tingting Shen, Yadong Mu
Computer Vision and Pattern Recognition (CVPR), 2026
arxiv / code / website /

ID-Crafter is a multi-subject video synthesis model that generates subject-consistent videos with multiple references.

PartCrafter: Structured 3D Mesh Generation via Compositional Latent Diffusion Transformers

Yuchen Lin*, Chenguo Lin*, Panwang Pan, Honglei Yan, Yiqiang Feng, Yadong Mu, Katerina Fragkiadaki
Neural Information Processing Systems (NeurIPS), 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.

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.

	ID-Crafter: VLM-Grounded Online RL for Compositional Multi-Subject Video Generation Panwang Pan, Jingjing Zhao, Yuchen Lin, Chenguo Lin, Chenxin Li, HengYu Liu, Tingting Shen, Yadong Mu Computer Vision and Pattern Recognition* (CVPR), 2026 arxiv / code / website / ID-Crafter is a multi-subject video synthesis model that generates subject-consistent videos with multiple references.
	MoVieS: Motion-Aware 4D Dynamic View Synthesis in One Second Chenguo Lin, Yuchen Lin, Panwang Pan, Yifan Yu, Honglei Yan, Katerina Fragkiadaki, Yadong Mu Computer Vision and Pattern Recognition (CVPR), 2026 arxiv / code / website / youtube / MoVieS is a feed-forward framework that jointly reconstructs appearance, geometry and motion for 4D scene perception from monocular videos.
	Diff4Splat: Controllable 4D Scene Generation with Latent Dynamic Reconstruction Models Panwang Pan, Chenguo Lin, Jingjing Zhao, Chenxin Li, Yuchen Lin, Kairun Wen, Yunlong Lin, Yixuan Yuan, Yadong Mu, Zhiwen Fan Computer Vision and Pattern Recognition (CVPR), 2026 arxiv / code / website / Diff4Splat is a generalizable framework for controllable 4D scene generation from a single image using a video diffusion model.
	PartCrafter: Structured 3D Mesh Generation via Compositional Latent Diffusion Transformers Yuchen Lin, Chenguo Lin, Panwang Pan, Honglei Yan, Yiqiang Feng, Yadong Mu, Katerina Fragkiadaki Neural Information Processing Systems (NeurIPS), 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.
	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.

Yuchen Lin | 林宇辰

News

Publications

ID-Crafter: VLM-Grounded Online RL for Compositional Multi-Subject Video Generation

PartCrafter: Structured 3D Mesh Generation via Compositional Latent Diffusion Transformers

OmniPhysGS: 3D Constitutive Gaussians for General Physics-based Dynamics Generation

ID-Crafter: VLM-Grounded Online RL for Compositional Multi-Subject Video Generation

MoVieS: Motion-Aware 4D Dynamic View Synthesis in One Second

Diff4Splat: Controllable 4D Scene Generation with Latent Dynamic Reconstruction Models

PartCrafter: Structured 3D Mesh Generation via Compositional Latent Diffusion Transformers

OmniPhysGS: 3D Constitutive Gaussians for General Physics-based Dynamics Generation

InstructLayout: Instruction-Driven 2D and 3D Layout Synthesis with Semantic Graph Prior

Experiences

Carnegie Mellon University

School of Computer Science

Machine Learning Department

Machine Intelligence Lab

Wangxuan Institute of Computer Technology, Peking University

Peking University

School of Electronics Engineering and Computer Science

Services

Selected Awards

Miscellaneous