Image of Jane Smith

Researcher & MSc Student
Computer Engineering

Faculty of Computer and Informatics
Maslak, 34469 Istanbul/Turkey
Room: 4105 HP Lab

I am an MSc student and graduate research assistant in Computer Engineering at Istanbul Technical University. I am also a researcher at ITU SIMIT Lab under the supervision of Hazım Kemal Ekenel.

My research interests are unsupervised learning, representation learning, biometrics, metric learning, zero-few shot learning, and explainability.

I received my BS in computer engineering from Istanbul Technical University in 2019. I completed my graduation project "Thermal-to-Visible Face Recognition" under the direction of Prof. Hazım Kemal Ekenel.

I work as a teaching assistant in the following courses BLG-513E (Image Processing grad course), BLG-506E (Computer Vision with Deep Learning grad course), ISE-309 (multimedia systems), BLG-252E (Object Oriented Programming), and BLG-374E (Technical Communication for Computer Engineers).

I have also have been a reviewer for ICIP (International Conference on Image Processing) and WACVMAPA (WACV2022 - Workshop on Manipulation, Adversarial, and Presentation Attacks in Biometrics).

I am working on multiple industrial and academic collaboration projects. You can see them below, and you can reach the papers on the publications page.

  • Video and photo Anomaly detection framework that combines multiple layers of representations with a one-class classification approach. It is joint work with researchers from ISTI-CNR, Italy.
  • Low-resolution face recognition project, which is supported by TUBITAK (Scientific and Technological Research Council of Turkey) and ARRS (Slovenian Research Agency. It is a collaboration with Assoc. Prof. Vitomir Struc from the University of Ljubljana.
  • Masked face detection and usage analysis with Twitter data to analyze the usage of masks for the COVID-19 pandemic. Joint project with QCRI (Qatar Computing Research Institute). We are organizing the FG workshop with a similar topic
  • Signature verification on real-world documents with Yapı Kredi Technologies. One of the biggest banks in Turkey uses the developed system to detect signature forgeries on banking transaction documents. We tried to analyze the applicability of the current state-of-the-art signature verification models on real-world banking documents. We developed a signature stamp cleaning model to increase the performance of the trained models. We presented our findings and contributions to the CVPR 2020 Biometrics Workshop.


  • M.S. in Computer Engineering with Computer Vision focus, 2019 - 2022 (Expected)
    Istanbul Technical University
  • B.S. in Computer Engineering, 2014 - 2019
    Istanbul Technical University