|
I am a final-year PhD candidate in computer vision and machine learning at the University of Amsterdam, supervised by Prof. Theo Gevers. I am part of the Delta Lab, a research collaboration with the Bosch Center for Artificial Intelligence, where I closely collaborate with Dr. Maxim Tatarchenko and Dr. Vien Ngo. My research interests lie in 3D generative models and 3D computer vision, particularly for robotics and AR/VR. I am currently spending this summer at Apple Zurich, working on digital humans. Last summer, I interned at Meta Reality Labs in Redmond, WA, working with Dr. Sho Nakagome. There, I focused on multimodal representation learning for AR/VR. Prior to that, I obtained my Master’s degree in Informatics from Technical University of Munich, where I worked on research projects at the Computer Vision Group led by Prof. Daniel Cremers, and completed my thesis on 3D reconstruction at the Smart Robotics Lab led by Prof. Stefan Leutenegger. I'm open to collaborations, feel free to reach out! Email / LinkedIn / Google Scholar / GitHub / X |
|
|
June, 2026: I joined Apple
June, 2025: I joined Meta Reality Labs February, 2025: I gave a talk at the Kickstart AI Meetup: AI in Interactive Worlds - All about Spatial Computing. July, 2024: Our work SceneTeller: Language-to-3D Scene Generation was accepted at ECCV 2024! The code is now available! April, 2024: I gave a talk on Learning Vision Based Autonomous Lateral Vehicle Control without Supervision with İdil Sülo to ATS Research group at Scania. January, 2023: Our work Learning Vision Based Autonomous Lateral Vehicle Control without Supervision was accepted at Applied Intelligence! September, 2022: I started my PhD in computer vision, under the supervision of Prof. Theo Gevers at the UvA-Bosch Delta Lab! March, 2022: I graduated with a Master's degree from Technical University of Munich! 🎓 |
as a research intern and will be based in Zurich, Switzerland for the next few months!
as a research scientist intern - I will be based in Redmond, WA for the upcoming months working with |
|
|
|
InStyle: Instant Appearance Stylization of 3D Shapes
Melis Öcal, Xiaoyan Xing, Yue Li, Vien Ngo, Sezer Karaoğlu, Theo Gevers MUSTCV Workshop, ECCV, 2026 (Oral) project page / arXiv InStyle is a diffusion-based feed-forward 3DGS style editor that uses geometry–appearance factorized latents and multi-view distillation, delivering edits in ~0.26 s while generalizing to unseen assets and real-world captures. |
|
GaussianTeller: Native 3D Gaussian Generation
Melis Öcal, Xiaoyan Xing, Vien Ngo, Sezer Karaoğlu, Theo Gevers CV4Metaverse Workshop, CVPR, 2025 paper GaussianTeller is a native 3D Gaussian splat generation framework that directly learns 3D diffusion priors from spatially-grouped Gaussians encoded into a structured latent space. |
|
SceneTeller: Language-to-3D Scene Generation
Melis Öcal, Maxim Tatarchenko, Sezer Karaoğlu, Theo Gevers ECCV, 2024 project page / arXiv Built using in-context learning, CAD model retrieval and 3DGS-based stylization, SceneTeller generates realistic and high-quality 3D spaces from natural language prompts. |
|
|
RealDiff: Real-world 3D Shape Completion using Self-Supervised Diffusion Models
Melis Öcal, Maxim Tatarchenko, Sezer Karaoğlu, Theo Gevers Curated Data for Efficient Learning Workshop, ECCV, 2026 project page / arXiv RealDiff formulates point cloud completion as a conditional generation problem directly on real-world measurements in a self-supervised way. To better deal with noisy observations, we leverage additional geometric cues. |
|
Learning Vision Based Autonomous Lateral Vehicle Control without Supervision
Qadeer Khan, İdil Sülo, Melis Öcal, Daniel Cremers Applied Intelligence, 2023 project page / paper A lateral vehicle control network can be trained from only an unlabeled sequence of images using novel view synthesis, without the need for a specialized setup on the car. |
|
|
Deep Learning Based 3D Reconstruction Using Images With Known Poses
Melis Öcal Master's Thesis Technical University of Munich, 2022 |
|
Reviewer |
|
Computer Vision 1, 2023-2024 |