About Me

I am a second year PhD student in Computer Science department at University of Maryland. Prior to that, I received my Bachelor's degree from Computer Engineering department at Bilkent University in Turkey with a minor in Mathematics. Currently I am working under the supervision of Prof. Rama Chellappa.



  • CMSC828T: Vision, Planning and Control for Aerial Robotics
  • CMSC828G: Image Understanding
  • CMSC726: Machine Learning
  • CMSC764: Advanced Numerical Optimization
  • CMSC661: Scientific Computing II
  • CMSC657: Introduction to Quantum Information Processing
  • ENEE620: Random Processes for Communications and Control
  • ENEE621: Estimation and Detection Theory

Teaching Assistant

  • CMSC131: Object Oriented Programming 1
  • CMSC426: Computer Vision


I am working in Deep Intermodal Video Analytics project as a research assistant. Currently my work is to generate spatio-temporal proposals for different activities happening in videos.


  • A Survey on Quantum Machine Learning

    A brief survey on quantum versions of machine learning algorithms from linear systems up to deep and reinforcement learning.

  • Vision, Planning and Control for AR Drone 2.0

    The project includes simulations and hardware implementation for vision, planning and control of AR Drone 2.0. Simulations were written in MATLAB, and hardware implementation was done with ROS with Python wrapper.

  • Human Activity Recognition Using LSTM

    An application written in Python, based on a CNN-LSTM model which recognizes human activities from videos.

    Code Report
  • CrypDist

    CrypDist aims to provide fast, safe and secure data management online. The data which is planned to be deal with is huge that can be expressed even in petabytes, and there is an abstract data for each data set which contains brief information about real data. In order to be capable of dealing with such big data, it will be distributed among multiple servers. Since whole data will not be used by a client, only abstracts of the data sets will be stored in client side. Clients can see the content of the real data and download whatever part of the data they need, manipulate it and upload it back to servers by the help of a decentralized database system called blockchain. For data safety, CrypDist will also have data backups to deal with arbitrary server failures. Also, to provide security for users’ data, the data to be stored will be encrypted not to allow third party users to access it. In addition, the connections between servers and clients should not leave a back door for unauthorized people.

    Code Report