Zhaopeng Cui

Research Professor, College of Computer Science and Technology, Zhejiang University

3D Vision Spatial Intelligence Embodied AI Physical Scene Understanding

I am a Research Professor at Zhejiang University, affiliated with the College of Computer Science and Technology and the State Key Laboratory of CAD&CG. My research lies at the intersection of computer vision, computer graphics, robotics, and machine learning.

My group works on 3D reconstruction, scene understanding, neural scene representations, SLAM, and physically grounded spatial perception. More broadly, I am interested in building systems that enable machines to perceive, reconstruct, and reason about the physical world at scale.

Before joining Zhejiang University, I was a Senior Researcher working with Prof. Marc Pollefeys at ETH Zurich. I received my Ph.D. in Computer Science from Simon Fraser University under the supervision of Prof. Ping Tan.

Email Google Scholar Full Publications

Research

My research lies at the intersection of computer vision, computer graphics, robotics, and machine learning. I am broadly interested in enabling machines to perceive, reconstruct, understand, and interact with the physical world through visual data.

3D Reconstruction

We study visual reconstruction methods that recover geometry, appearance, and structure of real-world scenes from sparse, incomplete, or casually captured observations, with an emphasis on scalable representations and editable modeling.

3D Scene Understanding

We aim to build holistic scene understanding systems that jointly reason about objects, layouts, occlusions, semantics, and spatial relations, bridging perception and structured 3D understanding in complex environments.

Physical-informed Modeling

We explore how physical principles can be integrated into visual modeling and generation, enabling scene representations that are not only geometrically plausible, but also predictive, controllable, and consistent with real-world dynamics.

Robot Perception and Planning

We investigate embodied perception systems for localization, mapping, and motion planning, with the goal of connecting 3D scene representations to robust decision-making and interaction in real environments.

Selected Publications

Representative publications organized by research directions.

3D Reconstruction

Free360: Layered Gaussian Splatting for Unbounded 360-Degree View Synthesis from Extremely Sparse and Unposed Views

Chong Bao, Xiyu Zhang, Zehao Yu, Jiale Shi, Guofeng Zhang, Songyou Peng, Zhaopeng Cui

CVPR 2025

A layered Gaussian splatting framework for sparse, unposed, and unbounded 360-degree scene reconstruction and view synthesis.

Project Paper

NeuMesh++: Towards Versatile and Efficient Volumetric Editing with Disentangled Neural Mesh-based Implicit Field

Chong Bao, Yuan Li, Bangbang Yang, Yujun Shen, Hujun Bao, Zhaopeng Cui, Yinda Zhang, Guofeng Zhang

TPAMI 2025

An extended neural mesh-based implicit representation for more versatile and efficient volumetric editing.

Project Paper

Learning Object-Compositional Neural Radiance Field for Editable Scene Rendering

Bangbang Yang, Yinda Zhang, Yinghao Xu, Yijin Li, Han Zhou, Hujun Bao, Guofeng Zhang, Zhaopeng Cui

ICCV 2021

An object-compositional NeRF framework that supports editable scene rendering with disentangled object-level control.

Project Paper

Global Structure-from-Motion by Similarity Averaging

Zhaopeng Cui, Ping Tan

ICCV 2015

A global structure-from-motion method based on similarity averaging for robust and scalable camera pose estimation.

Paper

3D Scene Understanding

InstaScene: Towards Complete 3D Instance Decomposition and Reconstruction from Cluttered Scenes

Zesong Yang, Bangbang Yang, Wenqi Dong, Chenxuan Cao, Liyuan Cui, Yuewen Ma, Zhaopeng Cui, Hujun Bao

ICCV 2025

A framework for decomposing cluttered scenes into complete instances while reconstructing intact object geometry and appearance.

Project Paper

Neural Rendering in a Room: Amodal 3D Understanding and Free-Viewpoint Rendering for the Closed Scene Composed of Pre-Captured Objects

Bangbang Yang, Yinda Zhang, Yijin Li, Zhaopeng Cui, Sean Fanello, Hujun Bao, Guofeng Zhang

SIGGRAPH 2022

A room-scale amodal 3D understanding and rendering framework for closed scenes composed of pre-captured objects.

Project Paper

DeepPanoContext: Panoramic 3D Scene Understanding with Holistic Scene Context Graph and Relation-based Optimization

Cheng Zhang, Zhaopeng Cui, Cai Chen, Shuaicheng Liu, Bing Zeng, Hujun Bao, Yinda Zhang

ICCV 2021

A panoramic scene understanding framework with holistic context graph modeling and relation-based optimization.

Project Paper

Holistic 3D scene understanding thumbnail

Holistic 3D Scene Understanding from a Single Image with Implicit Representation

Cheng Zhang, Zhaopeng Cui, Yinda Zhang, Bing Zeng, Marc Pollefeys, Shuaicheng Liu

CVPR 2021

A single-image 3D scene understanding framework based on implicit representation for holistic geometric reasoning.

Project Paper

Physical-informed Modeling

DiffWind: Physics-Informed Differentiable Modeling of Wind-Driven Object Dynamics

Yuanhang Lei, Boming Zhao, Zesong Yang, Xingxuan Li, Tao Cheng, Haocheng Peng, Ru Zhang, Yang Yang, Siyuan Huang, Yujun Shen, Ruizhen Hu, Hujun Bao, Zhaopeng Cui

ICLR 2026

A differentiable framework for modeling wind-object interaction from video with physics-informed simulation and reconstruction.

Project Paper

PhysSkin: Real-Time and Generalizable Physics-Based Animation via Self-Supervised Neural Skinning

Yuanhang Lei, Tao Cheng, Xingxuan Li, Boming Zhao, Siyuan Huang, Ruizhen Hu, Peter Yichen Chen, Hujun Bao, Zhaopeng Cui

CVPR 2026

A physics-informed self-supervised neural skinning framework for real-time and generalizable physics-based animation.

Project Paper

GaussianPrediction: Dynamic 3D Gaussian Prediction for Motion Extrapolation and Free View Synthesis

Boming Zhao, Yuan Li, Ziyu Sun, Lin Zeng, Yujun Shen, Rui Ma, Yinda Zhang, Hujun Bao, Zhaopeng Cui

SIGGRAPH 2024

A dynamic 3D Gaussian framework for future motion extrapolation and free-view synthesis in dynamic scenes.

Project Paper

Physics-guided reflection separation thumbnail

Physics-Guided Reflection Separation From a Pair of Unpolarized and Polarized Images

Youwei Lyu, Zhaopeng Cui, Si Li, Marc Pollefeys, Boxin Shi

TPAMI 2023

A physics-guided method for separating reflection using polarization cues from paired unpolarized and polarized images.

Paper

Robot Perception and Planning

NICE-SLAM: Neural Implicit Scalable Encoding for SLAM

Zihan Zhu, Songyou Peng, Viktor Larsson, Weiwei Xu, Hujun Bao, Zhaopeng Cui, Martin R. Oswald, Marc Pollefeys

CVPR 2022

A representative neural implicit SLAM framework for joint localization, mapping, and rendering.

Project Paper

PC-Planner: Physics-Constrained Self-Supervised Learning for Robust Neural Motion Planning with Shape-Aware Distance Function

Xujie Shen, Haocheng Peng, Zesong Yang, Juzhan Xu, Hujun Bao, Ruizhen Hu, Zhaopeng Cui

SIGGRAPH Asia 2024

A physics-constrained neural motion planning method with shape-aware distance fields for robust planning in complex environments.

Paper

DW-VIO: Deep Weighted Visual-Inertial Odometry

Guyuan Chen, Xiyue Guo, Xiaokun Pan, Yujun Shen, Guofeng Zhang, Hujun Bao, Zhaopeng Cui

IROS 2025

A deep visual-inertial odometry framework with learned weighting for robust state estimation.

Paper

CP-SLAM: Collaborative Neural Point-based SLAM System

Jiarui Hu, Mao Mao, Hujun Bao, Guofeng Zhang, Zhaopeng Cui

NeurIPS 2023

A collaborative neural point-based SLAM framework supporting multi-agent mapping and global refinement.

Project Paper

Full Publication List

News

Recent updates on papers, awards, service, and activities.

2026.03

Will be serving as Program Chair for 3DV 2028.

2026.02

Three papers accepted to CVPR 2026.

2026.01

One paper accepted to ICLR 2026.

2025.07

Five papers accepted to ICCV 2025.

2024.03

NICER-SLAM received the 3DV 2024 Best Paper Award Honorable Mention.

Awards, Service & Academic Activities

Selected Awards

3DV 2024 Best Paper Award Honorable Mention
IROS 2021 Best Paper Award Finalist on Safety, Security, and Rescue Robotics in memory of Motohiro Kisoi
ICRA 2020 Best Paper Award Finalist on Robot Vision

Professional Service

Associate Editor for Pattern Recognition, IEEE RA-L
Area Chair for CVPR, ECCV, IJCAI, 3DV
Associate Editor for ICRA, IROS

Openings

I am always looking for highly motivated Ph.D. students, research interns, and visiting students interested in 3D vision, embodied AI, scene understanding, SLAM, and physics-informed scene modeling. Please email me with your CV, transcript, and a short description of your research interests.