Tolga F. Aktas

Tolga F. Aktas

Ph.D. Candidate in Imaging Science @ Rochester Institute of Technology

Rochester Institute of Technology

Biography

I am a Ph.D. candidate in Carlson Center of Imaging Science at Rochester Institute of Technology. Supervised by Christopher Kanan, I am mainly interested in developing new methods in self-supervised learning, continual learning and computer vision. As part of my coursework at RIT, I have studied various topics across the spectrum of the imaging pipeline, and I was particularly interested in learning about Deep Learning for Vision. Noise & System Modeling, Statistical Machine Learning, Mathematics for Deep Learning, Multi-view Stereo Imaging.

I completed my Bachelor of Science at University of Rochester’s Electrical & Computer Engineering Department, as well as a minor in Computer Science. My relevant coursework includes Digital Image Processing, Computer Vision, Computer Audition, Stochastic Processes, Autonomous Mobile Robotics, Artificial Intelligence, Deep Learning, Machine Learning, Digital Signal Processing, Data Mining, Detection & Estimation Theory.

Download my resumé.

Education
  • Ph.D. Imaging Science

    Rochester Institute of Technology

  • B.S. Electrical & Computer Engineering, 2020

    University of Rochester

  • Minor: Computer Science, 2020

    University of Rochester

Skills

pytorch-icon
PyTorch

100%

Python

100%

cpp_logo
C++

90%

Experience

 
 
 
 
 
Microsoft
Applied Scientist Intern
Microsoft
Jun 2022 – Present Washington
  • Self-Supervised Learning Research on Cloud AI team.
 
 
 
 
 
kLab Rochester Institute of Technology
Graduate Researcher
Jan 2016 – Dec 2020 New York
Self-Supervised Learning, Continual Learning, Computer Vision.
 
 
 
 
 
Google
Software Engineering Intern
Google
Jun 2020 – Sep 2020 California
  • Google Geo 3D: High-Fidelity 3D Reconstruction from Satellite Imagery
  • Implemented a 3D reconstruction pipeline in C++ for camera parameter estimation and dense point cloud generation towards building a high-fidelity textured 3D reconstruction from satellite imagery.
  • Implemented image filtering algorithms in Python for texture processing, feature extraction and building contour detection, and super-resolution tasks on 30cm multi-spectral satellite imagery.
 
 
 
 
 
Qualcomm
Software Engineering Intern
Qualcomm
May 2019 – Sep 2020 California
  • VR Systems: High-Fidelity Realistic Avatar Rendering from User Expressions.
  • Built OpenGL ES application in C++ for avatar rendering using OBJ files developed in Autodesk Maya.
  • Implemented eye-tracking algorithm in C++ to integrate eye tracking capabilities
  • Implemented facial landmark detection and tracking algorithm in C++ to add real-time gesture tracking capabilities
  • Worked on variational/conditional VAE to generate avatar facial texture images using TensorFlow.
  • Investigated deep learning-based methods for alternative generation of occluded facial landmarks from speech and/or occluded image.

Accomplish­ments

Coursera
Deep Learning
See certificate
Modern C++ Development
Natural Language processing
Coursera
Structuring Machine Learning Projects
See certificate
Coursera
Convolutional Neural Networks
See certificate
Coursera
Sequence Models
See certificate
Deep Reinforcement Learning
Artificial Intelligence in Healthcare
Full-Stack Developer
Formulated informed blockchain models, hypotheses, and use cases.
See certificate
Coursera
Improving Deep Neural Networks: Hyperparameter Tuning, Regularizaton and Optimization
See certificate

Recent & Upcoming Talks

Recent Publications

Quickly discover relevant content by filtering publications.
(2013). An example conference paper. In ICW.

PDF Cite Project Slides

Popular Topics

Contact