Oğuzhan Fatih Kar

I am a Ph.D. student in Computer Science at the Swiss Federal Institute of Technology (EPFL), where I am advised by Amir Zamir. My current research interests are in building robust, adaptive, and multimodal perception systems that can scale and operate in the real world. My recent works on this topic are BRAVE, 4M & 4M-21, Rapid Network Adaptation, 3D Common Corruptions, and Cross-Domain Ensembles.

I received the B.S. and M.S. degrees in Electrical Engineering at METU, where I have worked with Figen Oktem. My M.S. work had focused on developing high-resolution and compressive reconstruction techniques for computational imaging.

Update: I am starting to look for research scientist positions. Feel free to reach out to discuss potential opportunities or collaborations.

Email  /  CV  /  Google Scholar  /  Github  /  LinkedIn  /  Twitter

Education

Ph.D. in Computer Science, 2024 (Expected)
EPFL

M.S. in Electrical and Electronics Engineering, 2019
METU (GPA:3.93/4.00) / Thesis

B.S. in Electrical and Electronics Engineering, 2017
METU (GPA:3.90/4.00)

News
  • June, 2024: I gave a talk at ETH Zurich on multimodal foundation models.
  • Nov, 2023: I started my student researcher position at Google Zurich! I'll be focusing on multimodal foundation models.
Recent Work
4M-21: An Any-to-Any Vision Model for Tens of Tasks and Modalities

O.F. Kar*, R. Bachmann*, D. Mizrahi*, A. Garjani, M. Gao, D. Griffiths, J. Hu, A. Dehghan, A. Zamir
arXiv, 2024
[Website] [Code] [Demo]

BRAVE: Broadening the visual encoding of vision-language models

O.F. Kar, A. Tonioni, P. Poklukar, A. Kulshrestha, A. Zamir, F. Tombari
ECCV, 2024
[Website]

Unraveling the Key Components of OOD Generalization via Diversification

H. Benoit*, L. Jiang*, A. Atanov*, O.F. Kar, M. Rigotti, A. Zamir
ICLR, 2024
[arXiv]

4M: Massively Multimodal Masked Modeling

D. Mizrahi*, R. Bachmann*, O.F. Kar, T. Yeo, M. Gao, A. Dehghan, A. Zamir
NeurIPS, 2023 [Spotlight, Top 4%]
[Website]

Rapid Network Adaptation: Learning to Adapt Neural Networks Using Test-Time Feedback

T. Yeo, O.F. Kar, Z. Sodagar, A. Zamir
ICCV, 2023
[Website]

3D Common Corruptions and Data Augmentation

O.F. Kar, T. Yeo, A. Atanov, A. Zamir
CVPR, 2022 [Oral, Top 4%]
[Website] [Code] [Video] [Live Demo] [TrustML Talk]

Robustness via Cross-domain Ensembles

O.F. Kar*, T. Yeo*, A. Zamir
ICCV, 2021 [Oral, Top 3%]
[Website] [Code] [Video] [Slides]

Robust Learning Through Cross-task Consistency

A. Zamir*, A. Sax*, T. Yeo, O.F. Kar, N. Cheerla, R. Suri, Z. Cao, J. Malik, L. Guibas
Arxiv, 2020. CVPR, 2020 [Best Paper Award Nominee, Oral]
[Live Demo] [Visuals] [Website] [Code] [ECCV 2020 Demo Video]

Misc

Robustness and uncertainty estimation for visual perception

This is a presentation I made on the relation between robustness and uncertainty in computer vision. The video also includes an overview of three important papers in uncertainty estimation in deep learning models.

[Slides] [Paper #1] [Paper #2] [Paper #3]

M.S. Work (2018-2021)

(Complete list on Google Scholar)

High-resolution Multi-spectral Imaging with Diffractive Lenses and Learned Reconstruction

F.S. Oktem, O.F. Kar, C. D. Bezek, F. Kamalabadi
IEEE Transactions on Computational Imaging, 2021
[Arxiv]

Compressive Spectral Imaging with Diffractive Lenses
O.F. Kar, F.S. Oktem
Optics Letters, 2019
[arXiv]

Real-time Compressive Video Reconstruction for Spatial Multiplexing Cameras

O.F. Kar, A. Gungor, H.E. Guven
IEEE GLOBALSIP, 2019
[Visuals]

Learning-based Regularization for Spatial Multiplexing Cameras
O.F. Kar, A. Gungor, H.E. Guven
IEEE GLOBALSIP, 2019

A Transform Learning-based Deconvolution Technique with Super-resolution and Microscanning Applications
A. Gungor*, O.F. Kar*
IEEE ICIP, 2019

A Matrix-free Reconstruction Method for Compressive Focal Plane Array Imaging
A. Gungor, O.F. Kar, H.E. Guven
IEEE ICIP, 2018


Template