CV

Education

Ph.D in Computer and Communication Sciences, École Polytechnique Fédérale de Lausanne (EPFL), Switzerland, 2018 - July 2024
- Thesis: Visual Saliency Prediction for Natural Images, Comics Panels, and Comics Pages
Bachelor of Science in Computer Engineering, Middle East Technical University (METU), Ankara, Turkey, 2013 - 2018 (CGPA: 3.92, Ranked 3rd out of 130)
Izmir Science High School, İzmir, Turkey, 2009 - 2013 (GPA: 93.15 / 100)

Work experience

Research Assistant @ EPFL, Image and Visual Representation Lab - September 2019 - July 2024
- Developed a data augmentation technique for saliency prediction using latent diffusion
- Incorporated temporal data into image saliency prediction
- Modeled object appearance and size dissimilarities for saliency prediction
Project Supervision @ EPFL - September 2019 - July 2024
- Supervised projects including eye tracking, face detection, and temporal saliency estimation
Teaching Assistant @ EPFL - January 2019 - July 2024
- Proposed and supervised projects for Computational Photography
- Conducted sessions for Introduction to Computation, Object-Oriented Programming, and Linear Algebra
Internship @ CERN - June 2017 - September 2017
- Created Python bindings for the C++ Data Access Library algorithms
Algorithmic Trading Project @ METU - September 2017 - July 2018
- Developed a web-based framework for designing and testing trading algorithms
Internship @ Turkish Aerospace Industries (TAI) - June 2016 - August 2016
- Implemented an auto-generated user interface in Groovy
Interdisciplinary Design Studio @ METU - September 2016 - June 2017
- Developed a body posture recognition and correction system for Autodesk

Skills

Theory: Computer Vision, Machine Learning, Generative AI, Transfer Learning, Deep Learning Model Architectures (CNN, RNN, GAN, VAE), Transformers, Diffusion Models, Image Processing, Image Manipulation, Feature Extraction, Linear Algebra, Probability and Statistics
Programming Languages: Python, C, C++, LaTeX, Java, SQL
Software Packages: PyTorch, OpenCV, scikit-learn, pandas, NumPy, SciPy, Jupyter, Seaborn
Deployment/Version Control: Conda, Docker, git, Kubernetes, Amazon Web Services (AWS)

Publications

Data Augmentation via Latent Diffusion for Saliency Prediction

Bahar Aydemir, Deblina Bhattacharjee, Tong Zhang, Mathieu Salzmann, Sabine Süsstrunk. (2024). "Data Augmentation via Latent Diffusion for Saliency Prediction." European Conference on Computer Vision (ECCV).

TempSAL - Uncovering Temporal Information for Deep Saliency Prediction

Bahar Aydemir, Ludo Hoffstetter, Tong Zhang, Mathieu Salzmann, Sabine Süsstrunk. (2023). "TempSAL - Uncovering Temporal Information for Deep Saliency Prediction." Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 6461-6470.

Modeling Object Dissimilarity for Deep Saliency Prediction

Bahar Aydemir, Deblina Bhattacharjee, Tong Zhang, Seungryong Kim, Mathieu Salzmann, Sabine Süsstrunk. (2022). " Modeling Object Dissimilarity for Deep Saliency Prediction." Transactions on Machine Learning Research (TMLR).