Ozan Ünal
I am a Senior Machine Learning and Computer Vision Research Scientist at Huawei, Zurich, where I lead the Image Restoration team. My team develops on-device, memory- and compute-efficient methods for image restoration and computational photography, including flicker debanding, demoireing, inpainting, and super-resolution for next-generation flagship camera pipelines. Before joining Huawei, I received my PhD from ETH Zurich under the supervision of Prof. Dr. Luc Van Gool. My doctoral research focused on data-efficient LiDAR semantic segmentation, with additional work on 3D visual grounding.
News
- BurstMamba is accepted to SIGGRAPH Asia 2026.
- My student’s work S4A is accepted to NeurIPS 2024 as a Spotlight.
- ConcreteNet and BST are accepted to ECCV 2024.
- DiAL receives the Best Paper Award — Honorable Mention from IEEE RA-L 2023.
- ConcreteNet wins the ICCV 2023 Object Localization Challenge.
- ScribbleKITTI is accepted to CVPR 2022 as an oral paper.
- Our follow-up work in MedIA 2021 receives the Runner-up Best Paper Award.
- Our paper in MICCAI 2019 is accepted as an oral presentation.
Highlighted papers
Bayesian Self-Training for Semi-Supervised 3D Segmentation
Ozan Unal, Christos Sakaridis, Luc Van Gool
Four Ways to Improve Verbo-visual Fusion for Dense 3D Visual Grounding
Ozan Unal, Christos Sakaridis, Suman Saha, Luc Van Gool
Runner-up Best Paper Award
Discwise Active Learning for LiDAR Semantic Segmentation
Ozan Unal, Dengxin Dai, Ali Tamer Unal, Luc Van Gool
ORAL
Scribble-Supervised LiDAR Semantic Segmentation
Ozan Unal, Dengxin Dai, Luc Van Gool
All papers
2026
-
Keyframe-Centric State-Space Modeling for Burst Image Super-ResolutionOzan Unal, Steven Marty, Dengxin Dai
2025
-
RealDriveSim: A Realistic Multi-Modal Multi-Task Synthetic Dataset for Autonomous DrivingArpit Jadon, Haoran Wang, Phillip Thomas, Michael Stanley, S. Nathaniel Cibik, Rachel Laurat, Omar Maher, Lukas Hoyer, Ozan Unal, Dengxin Dai
-
Language-Guided Instance-Aware Domain-Adaptive Panoptic SegmentationElham Amin Mansour, Ozan Unal, Suman Saha, Benjamin Bejar, Luc Van Gool
-
Fine-Grained Spatial and Verbal Losses for 3D Visual GroundingSombit Dey, Ozan Unal, Christos Sakaridis, Luc Van Gool
2024
-
Scribbles for All: Benchmarking Scribble Supervised Segmentation Across DatasetsWolfgang Boettcher, Lukas Hoyer, Ozan Unal, Jan Eric Lenssen, Bernt Schiele
-
Bayesian Self-Training for Semi-Supervised 3D SegmentationOzan Unal, Christos Sakaridis, Luc Van Gool -
Four Ways to Improve Verbo-visual Fusion for Dense 3D Visual GroundingOzan Unal, Christos Sakaridis, Suman Saha, Luc Van Gool -
2D Feature Distillation for Weakly- and Semi-Supervised 3D Semantic SegmentationOzan Unal, Dengxin Dai, Lukas Hoyer, Yigit Baran Can, Luc Van Gool
2023
-
LiDAR Meta Depth CompletionWolfgang Boettcher, Lukas Hoyer, Ozan Unal, Ke Li, Dengxin Dai
-
Discwise Active Learning for LiDAR Semantic SegmentationOzan Unal, Dengxin Dai, Ali Tamer Unal, Luc Van Gool
2022
-
Scribble-Supervised LiDAR Semantic SegmentationOzan Unal, Dengxin Dai, Luc Van Gool -
Understanding Birds-Eye View of Road Semantics using an Onboard CameraYigit Baran Can, Alexander Liniger, Ozan Unal, Danda Pani Paudel, Luc Van Gool
-
End-To-End Optimization of LiDAR Beam Configuration for 3D Object Detection and LocalizationNiclas Vodisch, Ozan Unal, Ke Li, Luc Van Gool, Dengxin Dai
2021
-
Inference on Homeostatic Belief PrecisionOzan Unal, Orhun Caner Eren, Göktuğ Alkan, Frederike H Petzschner, Yu Yao, Klaas Enno Stephan
-
Improving Point Cloud Semantic Segmentation by Learning 3D Object DetectionOzan Unal, Luc Van Gool, Dengxin Dai
-
Frequency-Dependent Attenuation Reconstruction with an Acoustic ReflectorRichard Rau, Ozan Unal, Dieter Schweizer, Valery Vishnevskiy, Orcun Goksel
2019
-
Attenuation Imaging with Pulse-Echo Ultrasound Based on an Acoustic ReflectorRichard Rau, Ozan Unal, Dieter Schweizer, Valery Vishnevskiy, Orcun Goksel
Patents
-
Wim Abbeloos, Ozan Unal, Christos Sakaridis, Luc Van GoolEuropean Patent Office · EP4607474A1 · 2025
-
Wim Abbeloos, Niclas Vödisch, Ozan Unal, Ke Li, Luc Van Gool, Dengxin DaiEuropean Patent Office · EP4198567A1 · 2021
Curriculum vitae
Experience
Huawei
Senior Machine Learning and Computer Vision Research Scientist
–presentMachine Learning and Computer Vision Research Scientist
–
Education
ETH Zurich
PhD in Machine Learning and Computer Vision
MSc in Electrical Engineering and Information Technology
BSc in Electrical Engineering and Information Technology