Bariscan Kurtkaya

Bariscan Kurtkaya

Koc University

Biography

I am an ambitious deep learning researcher, eager to explore the capabilities of generative deep learning models and self-supervised methods. My specific interest lies in generative models, particularly in the realm of 2D and 3D computer vision. Currently, I am pursuing graduate studies at Koc University, focusing on video editing using diffusion models. Additionally, I proudly serve as one of the two ambassadors of Hugging Face and Microsoft in Turkey. Beyond my academic pursuits, I am also a professional athlete, firmly believing in the collective strength of teamwork. This ethos profoundly influences my approach to research and collaboration, and I am actively seeking opportunities to contribute to interdisciplinary teams.

Interests
  • Generative Models
  • 2D & 3D Computer Vision
  • Self-supervised Representation Learning
Education
  • MSc in Computer Science and Engineering, Present

    Koc University

  • BSc in Electronics and Communication Engineering, 2023

    Yildiz Technical University

Experience

 
 
 
 
 
Virginia Tech - GemLab VT
Visiting Researcher
August 2023 – Present Virginia / USA
Conducted research on a novel, training-free, zero-shot video editing framework that integrates with pre-trained text- to-image diffusion models.
 
 
 
 
 
Georgia Institute of Technology - Rehg Lab
Researcher
August 2023 – December 2023 Atlanta / USA
Proposed novel lightweight image editing approach to harness the power of pre-trained text-to-image latent diffusion models to overcome inconsistency and incoherency in video editing.
 
 
 
 
 
Princeton University & Washington University St. Louis - Astromusers
Researcher
August 2022 – Present New Jersey / USA
Conducting a research on exoplanet observations utilizing the James Webb Space Telescope’s NIRCam instrument, implementing deep learning techniques for self-supervised exoplanet detection.
 
 
 
 
 
University of Milan - AISLab
Visiting Research Intern
June 2022 – September 2022 Milan / Italy
Research on the explainability of ‘Capsule Networks’ with the aim of improving the feature representation of images by understanding the ’dynamic routing’ notion and incorporating it with CNNs and Vision Transformers.
 
 
 
 
 
Ollang Media Technologies
Research and Development Engineer
July 2021 – July 2022 Istanbul / Turkey
Conducted research and implementations on Ollang’s Text-To-Speech and Transformer-based machine translation technology.
 
 
 
 
 
Yildiz Technical University - MedaLab
Undergraduate Research Assistant
October 2021 – June 2022 Istanbul / Turkey
Lead the research on Driver Assistance System which won a grant from Scientific and Technological Research Council of Turkey (TUBITAK), aiming to improve night driving safety with object detection and lane detection with deep learning and image processing.
 
 
 
 
 
A Group Companies
Software Engineer Intern
August 2020 – October 2020 Istanbul / Turkey
Undertook the unmasked people detection project, which includes Object Detection algorithms, due to the Covid-19 spread.
 
 
 
 
 
Yıldız Technical University IEEE CAS Team
Software Team Member
July 2018 – March 2019 Istanbul / Turkey
Contributed to the Semi-Autonomous Underwater Robot implementing the software for solving several computer vision tasks.

Accomplish­ments

KUIS AI Center - Koc University
KUIS AI Fellow
Hugging Face
Hugging Face Ambassador
Microsoft
Microsoft Ambassador

Recent Publications

(2023). RAVE: Randomized Noise Shuffling for Fast and Consistent Video Editing with Diffusion Models. arXiv preprint.

PDF Code Project Slides

(2023). Deep Learning Based Driver Assistance System. Electrica Journal.

PDF Code Dataset Slides