Oliver Hahn

I am an ELLIS PhD student at TU Darmstadt and NVIDIA supervised by Stefan Roth and Laura Leal-Taixé. My research focuses on generative and agentic world modeling and understanding for spatial intelligence. Previously, I worked on visual scene understanding without human supervision. I am fortunate to collaborate closely with Nikita Araslanov, Simone Schaub-Meyer, Christian Rupprecht, and Daniel Cremers.

I received a Master's in Computational Engineering with a focus on deep learning and computer vision from TU Darmstadt, including an exchange at Tongji University Shanghai. Previously, I obtained a Master's in Mechanical Engineering from TU Darmstadt.

News
Oct 26	LucidLyra preprint is out on arXiv!
Mar 26	I joined NVIDIA's Spatial Intelligence Lab (SIL) as a Research Scientist Intern!
Feb 26	Scene-Centric Unsupervised Video Panoptic Segmentation got accepted to CVPR 2026.
Jul 25	3 papers accepted! DFI-OmniStereo at IROS, MR-DINOSAUR at ICCVW, and SceneDINO at ICCV.
Mar 25	Scene-Centric Unsupervised Panoptic Segmentation got accepted as a CVPR 2025 Highlight paper!

Sep 24	Our paper Boosting Unsupervised Semantic Segmentation with Principal Mask Proposals got accepted at TMLR.
Jul 24	I will be attending the International Computer Vision Summer School (ICVSS) 2024!
Jun 24	Our paper won the best student paper award at the CVPR 2024 AIS Workshop.
Jun 23	Our paper Semantic Self-adaptation: Enhancing Generalization with a Single Sample got accepted at TMLR.
Oct 22	I started as a PhD student with Stefan Roth at the Visual Inference Lab at Technical University of Darmstadt!

Research

First author publications highlighted. Hover projects for animation.

	LucidLyra: Feed-Forward Generative 3D Scene Reconstruction and Joint Understanding via Multi-Task Distillation Oliver Hahn, Rodrigo Marcuzzi, Sangeetha Grama Srinivasan, Alessandro Burzio, Sherwin Bahmani, Jiahui Huang, Stefan Roth, Shengyu Huang, Qunjie Zhou, Laura Leal-Taixé Preprint 2026 Paper Project Page Code
	Contrastive Energy Fields for Inference-Time Procedure Planning in Instructional Videos Mohamed Afham, Christoph Reich, Oliver Hahn, Daniel Cremers, Stefan Roth GCPR 2026 (Best Paper Honorable Mention) Paper Project Page Code
	Scene-Centric Unsupervised Video Panoptic Segmentation Christoph Reich, Oliver Hahn, Nikita Araslanov, Laura Leal-Taixé, Christian Rupprecht, Daniel Cremers, Stefan Roth CVPR 2026 Paper Project Page Code
	Feed-Forward SceneDINO for Unsupervised Semantic Scene Completion Aleksandar Jevtić, Christoph Reich, Felix Wimbauer, Oliver Hahn, Christian Rupprecht, Stefan Roth, Daniel Cremers ICCV 2025 Paper Project Page Code
	Motion-Refined DINOSAUR for Unsupervised Multi-Object Discovery Xinrui Gong, Oliver Hahn, Christoph Reich, Krishnakant Singh, Simone Schaub-Meyer, Daniel Cremers, Stefan Roth ICCVW 2025 (Oral at ILR+G Workshop) Paper Code
	Scene-Centric Unsupervised Panoptic Segmentation Oliver Hahn, Christoph Reich, Nikita Araslanov, Daniel Cremers, Christian Rupprecht, Stefan Roth CVPR 2025 (Highlight) Paper Project Page Code
	Boosting Omnidirectional Stereo Matching with a Pre-trained Depth Foundation Model Jannik Endres, Oliver Hahn, Charles Corbière, Simone Schaub-Meyer, Stefan Roth, Alexandre Alahi IROS 2025 Paper Project Page Code
	Boosting Unsupervised Semantic Segmentation with Principal Mask Proposals Oliver Hahn, Nikita Araslanov, Simone Schaub-Meyer, Stefan Roth TMLR 2024 Paper Project Page Code
	A Perspective on Deep Vision Performance with Standard Image and Video Codecs Christoph Reich, Oliver Hahn, Daniel Cremers, Stefan Roth, Biplob Debnath CVPRW 2024 (Best Student Paper at AIS Workshop) Paper
	Semantic Self-adaptation: Enhancing Generalization with a Single Sample Sherwin Bahmani, Oliver Hahn, Eduard Zamfir, Nikita Araslanov, Daniel Cremers, Stefan Roth TMLR 2023* Paper Code

Academic Service

Reviewer: ECCV (since 2024), CVPR (since 2024, Outstanding Reviewer 2025), ICCV (since 2025), IROS (since 2025), TMLR (since 2024)

Teaching Assistant: Computer Vision I (Fall 2022), Computer Architecture (Spring 2024)