About
I am currently working as a PhD student at the Institute for Computational Science at the University of Zurich,supervised by Prof. Dr. Jan Dirk Wegner, and I am affiliated with the EcoVision Lab, located in ETH Zurich, since 2021. I received my Master’s degree in the field of Electrical and Electronics Engineering (EEE) at Middle East Technical University (METU), located in Ankara, Turkey as well as have been working in the ASELSAN Research Center as a research engineer to gain experience in topics such as computer vision and deep learning during my MS studies. My Master thesis focuses on fine-grained image retrieval via deep metric learning under the supervision of Prof. Dr. A. Aydın Alatan. Previously, I had received my two Bachelor of Science degrees with high honors in both the EEE and Physics departments from METU. My research interests include but are not limited to computer vision, deep metric learning, and remote sensing.
Please click here for my resume.
Education
Doctorate of Philosophy in University of Zurich, 2025 (expected)
- CGPA: 5.38/6.00
- Specialized in Data Science
- Thesis Topic: Deforestation Detection via SAR Image Time Series
- Advisor: Prof. Dr. Jan Dirk Wegner
Master of Science in Electrical Electronics Engineering at METU, 2021
- CGPA: 3.79/4.00
- Specialized in Signal Processing
- Thesis: Deep Metric Learning with Distance Sensitive Entangled Triplet Losses
- Advisor: Prof. Dr. A. Aydın Alatan
Bachelor of Science in Electrical Electronics Engineering at METU, 2017
- CGPA: 3.81/4.00
- Specialized in Bio-medical Engineering and Imaging
Double Major in Physics at METU, 2017
- CGPA: 3.67/4.00
- Specialized in Mathematical Physics and Relativity
Links
Publications
- Karaman, K., Garnot, V. S. F. & Wegner, J. D. (2023, September). Deforestation detection in the amazon with sentinel-1 sar image time series. ISPRS Geospatial Week 2023.
- Karaman, K., & Alatan, A. A. (2021, September). Metu loss: metric learning with entangled triplet unified loss. In 28th IEEE International Conference on Image Processing (ICIP). IEEE.
- Kayabasi, A., Karaman, K., & Akkaya, I. B. (2021, April). Comparison of distance metric learning methods against label noise for fine-grained recognition. In Automatic Target Recognition XXXI (Vol. 11729, p. 117290F). International Society for Optics and Photonics (SPIE).
- Akkaya I. B., & Karaman, K. (2020, May). A robust technique for real-time face verification with a generative network. In Real-Time Image Processing and Deep Learning (Vol. 11401, p. 1140107). International Society for Optics and Photonics (SPIE).
- Karaman, K., Akkaya I. B., Solmaz B., & Alatan A. A. (2020, October). A face recognition technique by representative learning with the quadruplets. In 28th Signal Processing and Communications Applications Conference (SIU). IEEE.
- Karaman, K., Akkaya I. B., & Alatan A. A. (2020, October). Metric learning with quadruplets on non-hierarchical labeled datasets. In 28th Signal Processing and Communications Applications Conference (SIU). IEEE.
- Karaman, K., Gundogdu, E., Koc, A., & Alatan, A. A. (2019, September). Quadruplet selection methods for deep embedding learning. In 26th IEEE International Conference on Image Processing (ICIP). IEEE.
- Karaman, K., & Akkaya I. B. (2019, October). Semi-supervised adversarial training of a lightweight neural network for visual recognition. In Counterterrorism, Crime Fighting, Forensics, and Surveillance Technologies II (Vol. 11166, p. 111660O). International Society for Optics and Photonics (SPIE).
- Solmaz, B., & Karaman, K. (2019, April). Modeling human activities via long short term memory networks. In 27th Signal Processing and Communications Applications Conference (SIU). IEEE.
- Karaman, K., Koc, A., & Alatan, A. A. (2018, October). Face recognition based on embedding learning. In Counterterrorism, Crime Fighting, Forensics, and Surveillance Technologies II (Vol. 10802, p. 108020J). International Society for Optics and Photonics (SPIE).
- Karaman, K., Gundogdu, E., Koc, A., & Alatan, A. A. (2018, May). A method for quadruplet sample selection in deep feature learning. In 26th Signal Processing and Communications Applications Conference (SIU). IEEE.
- Solmaz, B., Gundogdu, E., Karaman, K., Yucesoy, V., & Koc, A. (2017, October). Fine-grained visual marine vessel classification for coastal surveillance and defense applications. In Electro-Optical Remote Sensing XI (Vol. 10434, p. 104340A). International Society for Optics and Photonics (SPIE).