Bingxin Ke

Bingxin Ke (柯炳鑫)

I'm a third-year PhD student in Photogrammetry and Remote Sensing, ETH Zürich, supervised by Prof. Konrad Schindler . I obtained my Master's degree in Geomatics Engineering from ETH Zürich and Bachelor's degree from Wuhan University.

My research interests include computer vision (especially scene understanding and reconstruction), machine learning, and remote sensing. I love solving interesting problems!

Email: bingxin.ke[AT]geod.baug.ethz.ch

News

2025-06 Our paper Marigold-DC is accepted to ICCV 2025.
2025-06 I start as a research intern at NVIDIA Zürich, with Laura Leal-Taixe, Shengyu Huang, and Qunjie Zhou
2025-03 Our paper RollingDepth is accepted to CVPR 2025.
2024-09 Our paper BetterDepth is accepted to NeurIPS 2024
2024-06/07 Invited to give talks on Marigold at Stanford and Technion.
2024-03 Our paper Marigold is accepted to CVPR 2024 (Best paper award candidate).
2023-12 This page is online.

Research

A Unified Solution to Video Fusion: From Multi-Frame Learning to Benchmarking

Zixiang Zhao, Haowen Bai, Bingxin Ke, Yukun Cui Lilun Deng Yulun Zhang Kai Zhang Konrad Schindler

arXiv

Marigold: Affordable Adaptation of Diffusion-Based Image Generators for Image Analysis

Bingxin Ke, Kevin Qu, Tianfu Wang Nando Metzger, Shengyu Huang, Bo Li, Anton Obukhov, Konrad Schindler

arXiv

Marigold-DC: Zero-Shot Monocular Depth Completion with Guided Diffusion

Massimiliano Viola, Kevin Qu, Nando Metzger, Bingxin Ke, Alexander Becker, Konrad Schindler Anton Obukhov,

ICCV 2025

Video Depth without Video Models (RollingDepth)

Bingxin Ke, Dominik Narnhofer, Shengyu Huang, Lei Ke, Torben Peters, Katerina Fragkiadaki, Anton Obukhov, Konrad Schindler

CVPR 2025

Repurposing Diffusion-Based Image Generators for Monocular Depth Estimation (Marigold)

Bingxin Ke, Anton Obukhov, Shengyu Huang, Nando Metzger, Rodrigo Caye Daudt, Konrad Schindler

CVPR 2024 (Oral Presentation, Best Paper Award Candidate, top 0.2%)

BetterDepth: Plug-and-Play Diffusion Refiner for Zero-Shot Monocular Depth Estimation

Xiang Zhang, Bingxin Ke, Hayko Riemenschneider, Nando Metzger, Anton Obukhov, Markus Gross, Konrad Schindler, Christopher Schroers

NeurIPS 2024

Spot the Difference: Learned DSM Updating

Bingxin Ke, Corinne Stucker, Konrad Schindler

Wissenschaftlich-Technische Jahrestagung der DGPF, 2023, (3rd Prize at the Karl-Kraus Award)

ImpliCity: City Modeling from Satellite thumbnail with Deep Implicit Occupancy Fields

Corinne Stucker, Bingxin Ke, Yuanwen Yue, Shengyu Huang, Iro Armeni, Konrad Schindler

ISPRS Congress, 2022, (Best Young Author Award)

Experience

Research Intern, NVIDIA Zürich, June. 2025 - present
Ph.D. candidate, ETH Zurich, Sep. 2022 - present.
Research assistant, PRS, ETH Zurich, May. 2021 - Mar. 2022.
M.Sc in Geomatics, ETH Zurich, Sep. 2020 - Aug. 2022.
Summer School, Imperial College London, Jul. - Aug. 2019.
Research assistant, LIESMARS, Wuhan University, Oct. 2017 - Jun. 2019.
B.Eng in Geomatics Engineering, Wuhan University, Sep. 2016 - Jun. 2020.

Honors & Awards

Best Paper Award Candidate (Marigold), CVPR 2024.
3rd Prize at the Karl-Kraus Award. 2023.
Outstanding Graduate, School of Geodesy and Geomatics, Wuhan University, 2020.
“Chen Yongling” Scholarship, Wuhan University, 2018.
Merit Student of Wuhan University, Wuhan University, 2018.
“Guanghua” Scholarship, Wuhan University, 2017.
Merit Student of Wuhan University, Wuhan University, 2017.

Academic Service

Journal reviewer: T-IP; T-PAMI; Pattern Recognition
Conference reviewer: CVPR'25; SIGGRAPH'25

Teaching

Teaching assistant, Project Work Geospatial Engineering, ETH Zurich, 2024 spring & 2025 spring.
Teaching assistant, Geodetic Project Course, ETH Zurich, 2023 spring.
Teaching assistant, Image-based Mapping, ETH Zurich, 2022 fall.