Fangjinhua Wang (王方锦华)
I am a Postdoctoral reseacher at ETH Zurich, working with Prof. Marc Pollefeys. I obtained a PhD in Computer Science at ETH Zurich, supervised by Prof. Marc Pollefeys.
Previously, I obtained a Master's degree in Robotics at ETH zurich (with distinction) and a Bachelor's degree in Mechatronics Engineering at Zhejiang University (with distinction). I am the recipient of ETH Medal (ETH Zurich, 2021) and Chu Kochen Award (highest honor at Zhejiang University, 2017).
I am open to research collaboration. Feel free to contact me via email if you are interested in 3D vision and robotics.
I am on the job market.
Email  / 
Google Scholar  / 
Github  / 
Linkedin  / 
X(Twitter)
|
|
News
2025.07: DiffMVS is accepted to T-PAMI (my first T-PAMI paper)!
2025.07: I am officially Dr. sc. ETH!
2025.04: I am selected to participate in the Doctoral Consortium at CVPR 2025!
2025.04: 3 papers are accepted to CVPR2025 (2 posters + 1 highlight)!
|
Research Overview
My research is motivated by the goal of enabling embodied AI agents to perceive, understand and interact with the physical world as humans. During my PhD, I have been focusing on neural representations of the real world, including perception of spatial positioning information (visual localization), geometry (3D reconstruction) and appearance (neural rendering).
On the one hand, I focus on efficient algorithms that are computation-friendly for AR/VR/MR/robotics in general settings. On the other hand, I am interested in accurate representations for challenging settings, such as large-scale scenes, non-Lambertian surfaces, dynamics, and sparse inputs.
Currently, I am interested in using generative models to generate unlimited data for various downstream applications.
|
Publications
* denotes equal contribution. ✝ denotes corresponding author.
|
|
Lightweight and Accurate Multi-View Stereo With Confidence-Aware Diffusion Model
Fangjinhua Wang*,
Qingshan Xu*,
Yew-Soon Ong✝,
Marc Pollefeys✝
T-PAMI, 2025
Paper /
Code
|
|
Open-Vocabulary Functional 3D Scene Graphs for Real-World Indoor Spaces
Chenyangguang Zhang,
Alexandros Delitzas,
Fangjinhua Wang,
Ruida Zhang,
Xiangyang Ji,
Marc Pollefeys,
Francis Engelmann
CVPR, 2025   (Highlight)
Paper /
Project Page
|
|
R-SCoRe: Revisiting Scene Coordinate Regression for Robust Large-Scale Visual Localization
Xudong Jiang,
Fangjinhua Wang✝,
Silvano Galliani,
Christoph Vogel,
Marc Pollefeys
CVPR, 2025
Paper /
Code
|
|
DepthSplat: Connecting Gaussian Splatting and Depth
Haofei Xu,
Songyou Peng,
Fangjinhua Wang,
Hermann Blum,
Daniel Barath,
Andreas Geiger,
Marc Pollefeys
CVPR, 2025
Paper /
Project Page /
Code
|
|
Learning-based Multi-View Stereo: A Survey
Fangjinhua Wang*,
Qingtian Zhu*,
Di Chang*,
Quankai Gao,
Junlin Han,
Tong Zhang,
Richard Hartley,
Marc Pollefeys✝
In arXiv. Under review
Paper
|
|
UniSDF: Unifying Neural Representations for High-Fidelity 3D Reconstruction of Complex Scenes with Reflections
Fangjinhua Wang,
Marie-Julie Rakotosaona,
Michael Niemeyer,
Richard Szeliski,
Marc Pollefeys,
Federico Tombari
NeurIPS, 2024
Paper /
Project Page
|
|
HSR: Holistic 3D Human-Scene Reconstruction from Monocular Videos
Lixin Xue,
Chen Guo,
Chengwei Zheng,
Fangjinhua Wang,
Tianjian Jiang,
Hsuan-I Ho,
Manuel Kaufmann,
Jie Song,
Otmar Hilliges
ECCV, 2024
Paper /
Project Page /
Code
|
|
GLACE: Global Local Accelerated Coordinate Encoding
Fangjinhua Wang*,
Xudong Jiang*,
Silvano Galliani,
Christoph Vogel,
Marc Pollefeys
CVPR, 2024
Paper /
Project Page /
Code
|
|
VolRecon: Volume Rendering of Signed Ray Distance Functions for Generalizable Multi-View Reconstruction
Yufan Ren*,
Fangjinhua Wang*,
Tong Zhang,
Marc Pollefeys,
Sabine Süsstrunk
CVPR, 2023
Paper /
Project Page /
Code
|
|
IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo
Fangjinhua Wang,
Silvano Galliani,
Christoph Vogel,
Marc Pollefeys
CVPR, 2022
Paper /
Code
|
|
PatchmatchNet: Learned Multi-View Patchmatch Stereo
Fangjinhua Wang,
Silvano Galliani,
Christoph Vogel,
Pablo Speciale,
Marc Pollefeys
CVPR, 2021   (Oral Presentation)
Paper /
Code
|
|
Self-Calibrated Multi-Sensor Wearable for Hand Tracking and Modeling
Nikhil Bharadwaj Gosala*,
Fangjinhua Wang*,
Zhaopeng Cui,
Hanxue Liang,
Oliver Glauser,
Shihao Wu,
Olga Sorkine-Hornung
IEEE Transactions on Visualization and Computer Graphics, 2021
|
CVPR Doctoral Consortium, 2025
ETH Medal, ETH Zurich, 2021
Chu Kochen Award(Highest Honor at ZJU, 12 undergrates are awarded every year), Zhejiang University, 2017
National Scholarship, Zhejiang University, 2015&2016
|
Computer Vision, ETH Zurich, Autumn 2021, Autumn 2022, Autumn 2023, Autumn 2024
3D Vision, ETH Zurich, Spring 2023, Spring 2024
Introduction to Machine Learning, ETH Zurich, Spring 2021
|
Xudong Jiang (2023): Msc student at ETH Zurich
- Semester project: GLACE (CVPR'24)
- Master thesis: R-SCoRe (CVPR'25)
- → PhD student at ETH Zurich, advised by Marc Pollefeys
Chenyangguang Zhang (2024): Msc student at Tsinghua University
- Master thesis: OpenFunGraph (CVPR'25)
- → PhD student at ETH Zurich, advised by Marc Pollefeys
|
Google (Zurich, Switzerland), Student Researcher, 06.2023-11.2023
Meta (Redmond, US), AI Research Scientist Intern, 06.2022-09.2022
Microsoft Mixed Reality & AI Zurich Lab (Zurich, Switzerland), Research Intern, 10.2020-12.2020
|
Journal reviewer: T-PAMI, IJCV, TIP, TVCG, RA-L, TCSVT, CAG, Neural Computing
Conference reviewer: CVPR, NeurIPS, ICCV, ECCV, AAAI, ICLR, ICML, ICRA, IROS, ACMMM, ACCV
|
|