Shengyu Huang

Shengyu Huang 黄盛宇

I am a Research Scientist at Nvidia Dynamic Vision and Learning Group, part of Nvidia Spatial Intelligence Lab. I received my PhD from ETH Zürich in 2024, mainly advised by Konrad Schindler. Previously, I was an intern at NVIDIA and Google. I obtained my master degree in Geomatics Engineering from ETH Zürich. My undergraduate study was done in Shanghai, where I spent my first year in Mathematics but changed my direction and completed a bachelor in Surveying and Mapping Engineering at Tongji University.

Email / Google Scholar / Github / LinkedIn / Twitter

Publications

My research interest lies in 3D vision, machine learning and robotics. If you find my work interesting and would like to chat or want to do an internship with me, feel free to shoot me an e-mail!

* denotes equal contribution; ^# denotes mentored students.

	SimULi: Real-Time LiDAR and Camera Simulation with Unscented Transforms Haithem Turki, Qi Wu, Xin Kang, Janick Martinez Esturo, Shengyu Huang, Ruilong Li, Zan Gojcic, Riccardo de Lutio arxiv, 2025 project / paper
	Cosmos-Drive-Dreams: Scalable Synthetic Driving Data Generation with World Foundation Models Core contributor. I led the post-training of Cosmos WFM for Lidar generation including data curation, tokenizer and diffusion model training. White paper, 2025 project / paper / code
	Rectified Point Flow: Generic Point Cloud Pose Estimation Tao Sun, Liyuan Zhu, Shengyu Huang, Shuran Song, Iro Armeni NeurIPS, 2025, (spotlight) paper / code / project / bibtex
	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 PAMI, 2025 paper / code / project / bibtex
	ReStyle3D: Scene-level Appearance Transfer with Semantic Correspondences Liyuan Zhu, Shengqu Cai, Shengyu Huang, Gordon Wetzstein, Naji Khosravan, Iro Armeni Siggraph, 2025 paper / code / project / bibtex
	RollingDepth: Video Depth without Video Models Bingxin Ke, Dominik Narnhofer, Shengyu Huang, Lei Ke, Torben Peters, Katerina Fragkiadaki, Anton Obukhov, Konrad Schindler CVPR, 2025 paper / code / project / bibtex
	LoopSplat: Loop Closure by Registering 3D Gaussian Splats Liyuan Zhu, Yue Li, Erik sandström, Shengyu Huang, Konrad Schindler, Iro Armeni 3DV, 2025, (oral) paper / code / project / bibtex
	NSS: A Spatiotemporal Benchmark on 3D Point Cloud Registration Under Large Geometric and Temporal Change Tao Sun^#, Yan Hao^#, Shengyu Huang, Silvio Savarese, Konrad Schindler, Marc Pollefeys, Iro Armeni ISPRS Journal of Photogrammetry and Remote Sensing, 2025 paper / code(coming soon) / project Supervised student project for 3DV at ETH Zurich.
	DGInStyle: Domain-Generalizable Semantic Segmentation with Image Diffusion Models and Stylized Semantic Control Yuru Jia^#, Lukas Hoyer, Shengyu Huang, Tianfu Wang, Luc Van Gool, Konrad Schindler, Anton Obukhov ECCV, 2024 paper / project / code (coming soon) / bibtex Supervised master thesis at ETH Zurich.
	Repurposing Diffusion-Based Image Generators for Monocular Depth Estimation Bingxin Ke, Anton Obukhov, Shengyu Huang, Nando Metzger, Rodrigo Caye Daudt, Konrad Schindler CVPR, 2024, (oral, Best Paper Award Candidate) paper / code / project / bibtex
	Living Scenes: Multi-object Relocalization and Reconstruction in Changing 3D Environments Liyuan Zhu^#, Shengyu Huang, Konrad Schindler, Iro Armeni CVPR, 2024, (highlight) paper / code / project / bibtex Supervised master thesis at ETH Zurich.
	Dynamic LiDAR Re-simulation using Compositional Neural Fields Hanfeng Wu^#, Xingxing Zuo, Stefan Leutenegger, Or Litany, Konrad Schindler, Shengyu Huang CVPR, 2024, (highlight) CVPR Data-Driven Autonomous Driving Simulation Workshop, 2024, (oral) paper / code / project / bibtex Supervised master thesis at ETH Zurich.
	Neural LiDAR Fields for Novel View Synthesis Shengyu Huang, Zan Gojcic, Zian Wang, Francis Williams, Yoni Kasten, Sanja Fidler, Konrad Schindler, Or Litany ICCV, 2023 ICCV Neural Fields for Autonomous Driving and Robotics Workshop, 2023, (oral) paper / project / poster / bibtex
	Neural Fields meet Explicit Geometric Representations for Inverse Rendering of Urban Scenes Zian Wang, Tianchang Shen, Jun Gao, Shengyu Huang, Jacob Munkberg, Jon Hasselgren, Zan Gojcic, Wenzheng Chen, Sanja Fidler CVPR, 2023 paper / project / video/ bibtex
	DEFLOW: Self-supervised 3D Motion Estimation of Debris Flow Liyuan Zhu^#, Yuru Jia^#, Shengyu Huang, Nicholas Meyer, Andreas Wieser, Konrad Schindler, Jordan Aaron CVPR Workshop, 2023, (Best Paper Award) paper / code / project / poster/ bibtex Supervised semester project at ETH Zurich.
	Dynamic 3D Scene Analysis by Point Cloud Accumulation Shengyu Huang, Zan Gojcic, Jiahui Huang, Andreas Wieser, Konrad Schindler ECCV, 2022 paper / code / project / video/ poster/ bibtex
	ImpliCity: City Modeling from Satellite with Deep Implicit Occupancy Fields Corinne Stucker, Bingxin Ke^#, Yuanwen Yue^#, Shengyu Huang, Iro Armeni, Konrad Schindler ISPRS Congress, 2022, (Best Young Author Award) paper / code / bibtex Supervised semester project at ETH Zurich.
	PREDATOR: Registration of 3D Point Clouds with Low Overlap Shengyu Huang, Zan Gojcic, Mikhail Usvyatsov, Andreas Wieser, Konrad Schindler CVPR, 2021, (oral) paper / code / project / video/ poster/ talk(中文) / bibtex
	Indoor Scene Recognition in 3D Shengyu Huang, Mikhail Usvyatsov, Konrad Schindler IROS, 2020 paper / code / video / bibtex

Experience

Research Scientist, NVIDIA, Dec. 2024 - now

Ph.D. student, ETH Zurich, Oct. 2020 - Nov. 2024

Student Researcher, Google, July - December. 2023

Research Scientist intern, NVIDIA Toronto AI Lab, April - Dec. 2022

M.Sc. in Geomatics, ETH Zurich, Sep. 2018 - Aug. 2020

B.Eng. in Surveying and Mapping Enginnering, Tongji University, Sep. 2015 - June 2018

Freshman, Mathematics, Tongji University, Sep. 2014 - Sep. 2015

Academic Service

Workshop organizer: Neural Fields Beyond Conventional Camera at ECCV'24

Conference reviewer: Siggraph, CVPR, ICCV, ECCV, 3DV, IROS

Journal reviewer: ToG, T-PAMI, RA-L, ISPRS Journal

Misc

I was a die-hard Lakers fan, Kobe Bryant inspires me the most so far. I think 罗大佑 is the greatest Chinese singer. Besides research / work, I like running, hiking, skiing and (analog) photography. I am a proud owner of a Nikon F3, Hasselblad Xpan, Leica M3 Double Stroke and a Contax G1 (green label).

Last update: April. 2025. Thanks for the template.