Mulin Yu (original) (raw)

Researcher · Shanghai AI Lab

I work on 3D vision, scene reconstruction, neural rendering, and structured geometric modeling. My recent research focuses on building digital worlds with both geometric precision and visual fidelity.

Portrait of Mulin Yu

About

I am currently a Researcher at Shanghai AI Lab. Before that, I was a Postdoctoral Researcher in the IDC group at Shanghai AI Lab, working closely with Bo Dai and Dahua Lin. I received my PhD from INRIA in 2022 under the supervision of Prof. Florent Lafarge.

My work spans surface reconstruction, city modeling, novel view synthesis, and geometry-aware representations for realistic 3D understanding.

Research Interests

ARTDECO: Towards Efficient and High-Fidelity On-the-Fly 3D Reconstruction with Structured Scene Representation

Guanghao Li*, Kerui Ren*, Lining Xu, Zhewen Zheng, Changjian Jiang, Xin Gao, Bo Dai, Jian Pu†, Mulin Yu†, Jiangmiao Pang

ICLR 2026

PLANING: A Loosely Coupled Triangle-Gaussian Framework for Streaming 3D Reconstruction

Changjian Jiang*, Kerui Ren*, Xudong Li, Kaiwen Song, Guanghao Li, Linning Xu, Tao Lu, Junting Dong, Yu Zhang†, Bo Dai, Mulin Yu†

arXiv 2026

AnySplat: Feed-forward 3D Gaussian Splatting from Unconstrained Views

Lihan Jiang*, Yucheng Mao*, Linning Xu, Tao Lu, Kerui Ren, Yichen Jin, Xudong Xu, Mulin Yu, Jiangmiao Pang, Feng Zhao, Dahua Lin, Bo Dai

ACM TOG 2025

Horizon-GS: Unified 3D Gaussian Splatting for Large-Scale Aerial-to-Ground Scenes

CVPR 2025

ObjectGS: Object-aware Scene Reconstruction and Scene Understanding via Gaussian Splatting

Ruijie Zhu, Mulin Yu, Linning Xu, Lihan Jiang, Yixuan Li, Tianzhu Zhang, Jiangmiao Pang, Bo Dai

ICCV 2025

HaloGS: Loose Coupling of Compact Geometry and Gaussian Splats for 3D Scenes

Changjian Jiang*, Kerui Ren*, Linning Xu, Jiong Chen, Jiangmiao Pang, Yu Zhang, Bo Dai, Mulin Yu

arXiv 2025

MV-CoLight: Efficient Object Compositing with Consistent Lighting and Shadow Generation

Kerui Ren, Jiayang Bai, Linning Xu, Lihan Jiang, Jiangmiao Pang, Mulin Yu†, Bo Dai†

NeurIPS 2025

Octree-GS: Towards Consistent Real-time Rendering with LOD-Structured 3D Gaussians

Kerui Ren*, Lihan Jiang*, Tao Lu, Mulin Yu, Linning Xu, Zhangkai Ni, Bo Dai

TPAMI 2025

Scaffold-GS: Structured 3D Gaussians for View-Adaptive Rendering

Tao Lu*, Mulin Yu*, Linning Xu, Yuanbo Xiangli, Limin Wang, Dahua Lin, Bo Dai

CVPR 2024 (Highlight)

Inter3D: A Benchmark and Strong Baseline for Human-Interactive 3D Object Reconstruction

Gan Chen, Ying He, Mulin Yu, F. Richard Yu, Gang Xu, Fei Ma, Ming Li, Guang Zhou

IJCAI 2025

GSDF: 3DGS Meets SDF for Improved Neural Rendering and Reconstruction

Mulin Yu*, Tao Lu*, Linning Xu, Lihan Jiang, Yuanbo Xiangli, Bo Dai

NeurIPS 2024

Sharp Feature Consolidation from Raw 3D Point Clouds via Displacement Learning

Tong Zhao, Mulin Yu, Pierre Alliez, Florent Lafarge

CAGD 2023

Repairing Geometric Errors in 3D Urban Models with Kinetic Data Structures

Mulin Yu, Florent Lafarge, Sven Oesau, Bruno Hilaire

ISPRS Journal of Photogrammetry and Remote Sensing 2022

City-scale 3D reconstruction system

A series of large-scale city reconstruction systems evolving from 1.0 to 3.0, progressively improving scalability, fidelity, and moving toward a unified city-scale 3D system for reconstruction, rendering, and editing.

Versions 1.0 / 2.0 / 3.0

Interactive simulation assets

A Real-to-Sim-to-Real embodied intelligence framework that enables closed-loop learning across real and simulated environments, addressing real-world data scarcity and significantly improving training efficiency for autonomous agents.

Reconstruction, editable assets, and simulation-ready worlds

Recognition

Reviewing

Reviewer for CVPR, ECCV, TPAMI, CAD, JPRS, SIGGRAPH, and related venues.

Research Focus

Surface reconstruction, city modeling, novel view synthesis, and geometry-aware 3D representations.