Dr. Oğuzhan Ulucan

Contact

Institut für Mathematik und Informatik
Walther-Rathenau-Str. 47
17489 Greifswald

Telefon: +49 3834 420 4623
Email: oguzhan.ulucan[at]uni-greifswald.de

Google Scholar  ResearchGate  LinkedIn

See Chair of Computer Science

Teaching (Tutorials)

Introduction to Computer Science
WiSe 2021/22, 2022/23, 2023/24, 2024/25, 2025/26

Software Engineering Internship
SoSe 2022, 2024, 2025, 2026

Evolutionary Algorithms
SoSe 2023

Short Biography

Oguzhan Ulucan is a postdoctoral researcher at the University of Greifswald, Germany (since 2025). He completed his Ph.D. in 2025 under the supervision of Prof. Dr. Marc Ebner, with a dissertation on computational color constancy and the perception of color illusions, for which he was nominated for the GI Dissertation Award. He received his B.Sc. (with honors) and M.Sc. in Electrical and Electronics Engineering from Izmir University of Economics, Turkey, in 2017 and 2020. His research interests include computer vision and machine learning, with a focus on color constancy, color illusions, intrinsic image decomposition, and low-light image enhancement.

News!

2026: My thesis is nominated for the GI Dissertation Award! I have presented it at Schloss Dagstuhl!

2026: Our paper is published in The International Journal of Computer Vision! 

2025: I have successfully defended my Ph.D. thesis! Here you can reach the work!

2025: Our work “Low-Light Image Enhancement through Multi-Scale Local Space Average Color” is accepted at EUSIPCO 2025, Palermo, Italy!

2025: Our work “Towards Explainable Hate Speech Detection” is accepted at Findings of ACL 2025, Vienna, Austria! 

2025: Our work “Challenges and Applications of Intrinsic Image Decomposition: A Short Review” which is related to the fundamentals of the intrinsic image decomposition is now accessible at SN Computer Science!

2024: Our work "A computational model for color assimilation illusions and color constancy" is accepted at ACCV 2024 (oral presentation acceptance rate: 5.6%)! Our work is chosen as an award candidate!

2024: I have been selected as a reviewer for the IEEE SPS Multimedia Signal Processing Technical Committee! I am looking forward to contributing to the SPS community!

2024: Our work “A Scale-space Approach for Surface Normal Vector Estimation from Depth Maps” is now accessible at SN Computer Science!

2024: Two papers are accepted at CCIW 2024, Milano, Italy!

  1. Revisiting Color Constancy Using CNNs: Including Recent Observations
  2. Intrinsic Image Decomposition based on Retinex Theory, Superpixel Segmentation and Scale-Space Computations

2023: Our work "Investigating Color Illusions from the Perspective of Computational Color Constancy" is accepted at VISAPP 2024, Rome, Italy!

2023: Our paper is accepted for publication in The International Journal of Computer Graphics - The Visual Computer!

Recent Refereed Publications

* Please note that all downloadable papers available here are preprints or open-access publications. For a complete list of my works, please refer to my Google Scholar page.

  • A Traditional Approach for Color Constancy and Color Assimilation Illusions with Its Applications to Low-Light Image Enhancement [Paper]
    Oguzhan Ulucan, Diclehan Ulucan, Marc Ebner
    International Journal of Computer Vision, 2026
  • Challenges and Applications of Intrinsic Image Decomposition: A Short Review [Paper]
    Diclehan Ulucan, Oguzhan Ulucan, Marc Ebner
    SN Computer Science, 2025
  • A Scale-space Approach for Surface Normal Vector Estimation from Depth Maps [Paper]
    Diclehan Ulucan, Oguzhan Ulucan, Marc Ebner
    SN Computer Science, 2024
  • Multi-scale color constancy based on salient varying local spatial statistics  [ Paper ]
    Oguzhan Ulucan, Diclehan Ulucan, Marc Ebner
    The Visual Computer, 2023
  • Low-Light Image Enhancement based on Intrinsic Image Decomposition [will appear] 
    Diclehan Ulucan, Oguzhan Ulucan, Marc Ebner
    VCIP 2025, Klagenfurt, Austria
  • Towards Explainable Hate Speech Detection [ Paper ] 
    Happy Khairunnisa Sariyanto, Diclehan Ulucan, Oguzhan Ulucan, Marc Ebner
    Findings of ACL 2025, Vienna, Austria
  • Low-Light Image Enhancement through Multi-Scale Local Space Average Color [ Paper ] 
    Oguzhan Ulucan, Diclehan Ulucan, Marc Ebner
    EUSIPCO 2025, Palermo, Italy
  • A Computational Model for Color Assimilation Illusions and Color Constancy (Best Paper Award Candidate) [ Paper ] 
    Oguzhan Ulucan, Diclehan Ulucan, Marc Ebner
    ACCV 2024, Hanoi, Vietnam
  • Revisiting Color Constancy Using CNNs: Including Recent Observations  [ Paper ] 
    Oguzhan Ulucan, Diclehan Ulucan, Marc Ebner
    CCIW 2024, Milano, Italy
  • Intrinsic Image Decomposition based on Retinex Theory, Superpixel Segmentation and Scale-Space Computations   [ Paper ] 
    Diclehan Ulucan, Oguzhan Ulucan, Marc Ebner
    CCIW 2024, Milano, Italy
  • Investigating Color Illusions from the Perspective of Computational Color Constancy  [ Paper ] 
    Oguzhan Ulucan, Diclehan Ulucan, Marc Ebner
    VISAPP 2024, Rome, Italy
  • Multi-scale Block-based Color Constancy  [ Paper ] 
    Oguzhan Ulucan, Diclehan Ulucan, Marc Ebner
    EUSIPCO 2023, Helsinki, Finland
  • CC-NORD: A Camera-Inveriant Global Color Constancy Dataset  [ Paper ] 
    Diclehan Ulucan, Oguzhan Ulucan, Marc Ebner
    EUSIPCO 2023, Helsinki, Finland
  • Block-based Color Constancy: The Deviation of the Salient Pixels  [ Paper ] [ Poster ]
    Oguzhan Ulucan, Diclehan Ulucan, Marc Ebner
    IEEE ICASSP 2023, Rhodes Island, Greece
  • Multi-scale Surface Normal Estimation from Depth Maps (Best Student Paper Award Candidate) [Paper] 
    Diclehan Ulucan, Oguzhan Ulucan, Marc Ebner
    IMPROVE 2023, Prague, Czech Republic
  • Intrinsic Image Decomposition: Challenges and New Perspectives (Best Student Paper Award Candidate) [Paper] 
    Diclehan Ulucan, Oguzhan Ulucan, Marc Ebner
    IMPROVE 2023, Prague, Czech Republic
  • BIO-CC: Biologically Inspired Color Constancy  [ Paper ] [ Poster ]
    Oguzhan Ulucan, Diclehan Ulucan, Marc Ebner
    BMVC 2022, London, UK
  • Color Constancy Beyond Standard Illuminants  [ Paper ] [ Poster ]
    Oguzhan Ulucan, Diclehan Ulucan, Marc Ebner
    IEEE ICIP 2022, Bordeaux, France
  • IID-NORD: A Comprehensive Intrinsic Image Decomposition Dataset [Paper] 
    Diclehan Ulucan, Oguzhan Ulucan, Marc Ebner
    IEEE ICIP 2022, Bordeaux, France 

Academic Services

  • London Imaging Meeting (LIM 2026), 22-24 June, London, UK, 2026.
  • International Conference on Pattern Recognition (ICPR 2026), 17 - 22 August, Lyon, France, 2026.
  • IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2026), 4-8 May, Barcelona, Spain, 2026.
  • 21st International Conference on Computer Vision Theory and Applications (VISAPP 2026), March 09 - 11, Marbella, Spain, 2026.
  • IEEE International Joint Conference on Neural Networks (IJCNN 2025), 30 June - 5 July, Rome, Italy, 2025.
  • IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2025), 06 - 11 April, Hyderabad, India, 2025.
  • International Conference on Pattern Recognition (ICPR 2024), 01 - 05 April, Koltaka, India, 2024.
  • IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2024), 14 - 19 April, Seoul, Korea, 2024.
  • International Conference on Advances in Image Processing (ICAIP 2024), 18 - 20 October, Chengdu, China, 2024.
  • International Conference on Advances in Image Processing (ICAIP 2023), 17 - 19 November, Beijing, China, 2023.
  • 8th International Conference on Computer Graphics and Virtuality (ICCGV 2025) 21 – 23 February 2025, Chengdu, China.
  • 8th International Conference on Advances in Image Processing (ICAIP 2024), 18 - 20 October 2024, Chengdu, China.
  • 7th International Conference on Advances in Image Processing (ICAIP 2023), 17 - 19 November 2023, Beijing, China.

Color Assimilation Illusions

In computer vision, we typically focus on the cases where the human visual system succeeds, such as color constancy. However, I believe that studying the failure cases of human perception alongside the successful ones can lead to more robust and explainable methods. This is why color assimilation illusions are central to my research: they are situations where our visual system is deceived, and analyzing them together with color constancy provides valuable insights for both phenomena.

In our previous studies, we have shown that this perspective leads to a learning-free color constancy method that performs competitively in both single- and multi-illuminant scenarios. More interestingly, the same method also extends to low-light image enhancement, a completely different computer vision task, which suggests that mechanisms inspired by the human visual system can serve as a basis for simple and explainable algorithms across multiple tasks.

Below are some examples of color assimilation illusions I have designed. In these illusions, the colors of the target regions (e.g., the triangles below) are identical, yet they are perceived as different due to their context, i.e., the local surroundings. If you are interested in using these illusions in your work, please feel free to contact me.