Yağız Nalçakan, PhD

Research Professor at Yonsei University

Yağız Nalçakan

I am a Research Professor at the Seamless Trans-X Lab (STL), Yonsei University, where I work with Prof. Shiho Kim on multispectral imaging and robust visual perception. My research sits at the intersection of computer vision, sensor fusion, and multimodal representation learning, with applications spanning intelligent mobility, robotics, and industrial and environmental sensing.

I received my Ph.D. in Computer Science from the Izmir Institute of Technology, where my dissertation focused on vehicle maneuver prediction for advanced driver assistance systems (ADAS) under the supervision of Prof. Dr. Yalin Bastanlar, in collaboration with TTTech Auto Turkey. During my doctoral studies, I spent a term as a visiting researcher at Seoul National University’s Vehicle Dynamics and Control Laboratory (VDCL) and the Future Mobility Technology Center (FMTC), supported by a research scholarship from TUBITAK.

My research interests center on computer vision and deep learning, with emphasis on multispectral camera systems spanning the visible, near-infrared, short-wave infrared, and long-wave infrared bands, perception under adverse conditions, representation learning, and vision-language modeling.

Research focus

  • Multispectral perception: Designing datasets and architectures that fuse RGB, NIR, SWIR, and LWIR imagery for detection, segmentation, and depth estimation.
  • Multimodal foundation models and representation learning: Extending self-supervised vision foundation models beyond RGB, and studying cross-spectral translation and domain adaptation under raw-sensor domain shift.
  • Vision-language and explainable perception: Exploring how multimodal models can interpret challenging scenes and make their reasoning inspectable.

News

  1. Our new preprint “SpectraDINO: Modality-Conditioned Adaptation of RGB Vision Foundation Models Across Infrared Bands” is now available on arXiv.
  2. Our paper “RPT-SR: Regional Prior attention Transformer for infrared image Super-Resolution” was published in the WACV 2026 proceedings.
  3. Our paper “SpectraDiff: Enhancing the fidelity of Infrared Image Translation with object-aware diffusion” was published in Computer Vision and Image Understanding.
  4. Our paper “RASMD: RGB and SWIR multispectral driving dataset for robust perception in adverse conditions” was published in Information Fusion.
  5. Our work “IVIFormer: Illumination-Aware Infrared-Visible Image Fusion via Adaptive Domain-Switching Cross Attention”, has been published on ICCV 2025 Workshop Proceedings.
  6. I organized and served as the Chair of the 2nd Workshop on Multispectral Imaging for Robotics and Automation (MIRA) at the International Conference on Computer Vision (ICCV) 2025.
  7. We publicly release the RASMD dataset, a large-scale RGB–SWIR benchmark for multispectral perception in adverse weather, available on arXiv and Huggingface.
  8. Our paper “Short-Wave Infrared (SWIR) Imaging for Robust Material Classification: Overcoming Limitations of Visible Spectrum Data” was published in Applied Sciences.

Selected publications

  1. SpectraDINO: Modality-Conditioned Adaptation of RGB Vision Foundation Models Across Infrared Bands
    SpectraDINO: Modality-Conditioned Adaptation of RGB Vision Foundation Models Across Infrared Bands
    Yagiz Nalcakan, Hyeongjin Ju, Incheol Park, Sanghyeop Yeo, Youngwan Jin, and Shiho Kim
    arXiv preprint arXiv:2605.02258 2026
    arXiv Bib Link CodearXiv
  2. RPT-SR: Regional Prior attention Transformer for infrared image Super-Resolution
    RPT-SR: Regional Prior attention Transformer for infrared image Super-Resolution
    Youngwan Jin, Incheol Park,  Yagiz Nalcakan, Hyeongjin Ju, Sanghyeop Yeo, and Shiho Kim
    In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) 2026
    Bib Link CodeWACV
  3. SpectraDiff: Enhancing the fidelity of Infrared Image Translation with object-aware diffusion
    SpectraDiff: Enhancing the fidelity of Infrared Image Translation with object-aware diffusion
    Incheol Park, Youngwan Jin,  Yagiz Nalcakan, Hyeongjin Ju, Sanghyeop Yeo, and Shiho Kim
    Computer Vision and Image Understanding 2026
    Bib Link Code WebsiteCVIU
  4. RASMD: RGB and SWIR multispectral driving dataset for robust perception in adverse conditions
    RASMD: RGB and SWIR multispectral driving dataset for robust perception in adverse conditions
    Youngwan Jin, Michal Kovac,  Yagiz Nalcakan, Incheol Park, Sanghyeop Yeo, Hyeongjin Ju, and Shiho Kim
    Information Fusion 2026
    arXiv Bib Link WebsiteINFFUS
  5. IVIFormer: Illumination-Aware Infrared-Visible Image Fusion via Adaptive Domain-Switching Cross Attention
    IVIFormer: Illumination-Aware Infrared-Visible Image Fusion via Adaptive Domain-Switching Cross Attention
    Incheol Park, Youngwan Jin,  Yagiz Nalcakan, Hyeongjin Ju, Sanghyeop Yeo, and Shiho Kim
    In Proceedings of the IEEE/CVF International Conference on Computer Vision Workshops (ICCVW) 2025
    Bib LinkICCVw
  6. Pix2Next: Leveraging Vision Foundation Models for RGB to NIR Image Translation
    Pix2Next: Leveraging Vision Foundation Models for RGB to NIR Image Translation
    Youngwan Jin, Incheol Park, Hanbin Song, Hyeongjin Ju,  Yagiz Nalcakan, and Shiho Kim
    Technologies 2025
    arXiv Bib Link Code WebsiteTech

View all publications >>

Service

Editorial reviewing and community roles.

Journal reviewer

  • Elsevier Information Fusion (INFFUS)
  • Elsevier Computer Vision and Image Understanding (CVIU)
  • Elsevier Expert Systems with Applications (ESWA)
  • IEEE Transactions on Intelligent Vehicles (T-IV)
  • IEEE Transactions on Intelligent Transportation Systems (T-ITS)

Conference reviewer

  • IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
  • IEEE/CVF International Conference on Computer Vision (ICCV)
  • AAAI Conference on Artificial Intelligence (AAAI)
  • IEEE International Conference on Intelligent Transportation Systems (ITSC)
  • IEEE Intelligent Vehicles Symposium (IV)

Program committee