Vassilis Athitsos Home Page

Vassilis Athitsos
Professor
Computer Science and Engineering Department
University of Texas at Arlington
Office: ERB 623

Publications
Teaching
Curriculum Vitae
Contact Info

Office Hours (via Microsoft Teams): MW 11:00am - 12:30pm.

For the Spring 2023 semester, office hours are conducted via Microsoft Teams. During office hours:

You can request a meeting via e-mail or via chat on Microsoft Teams.
I will acknowledge the message as soon as I can.
I will be calling people on Teams in the order in which I receive their requests.

Teaching

Spring 2023: CSE 4392 - Special Topics: Neural Networks and Deep Learning.

Fall 2022: CSE 4309 - Machine Learning.

Spring 2021: CSE 4310 - Introduction to Computer Vision.

Courses taught in the past.

Research - Publications - VLM Research Lab

My main research areas are computer vision, machine learning, and data mining. A large part of my current work focuses on gesture and sign language recognition, human motion analysis, and similarity-based retrieval in image, video and multimedia databases. Details on the various projects I've been working on are provided in my publications. Brief descriptions of selected ongoing and past projects can be found at the following links:

Assessing cognitive skills in children through performance in physical and computer-based tasks
Developing computer vision and data mining methods for assessing cognitive skills in children, based on the children's performance in structured physical and computer-based tasks.
ASL dictionary search
Developing computer vision methods for a system that lets users search dictionaries of American Sign Language (ASL), so as to look up the meaning of unknown signs. Videos and annotations from this project are available here.
Subsequence matching in large databases of time series and strings.
The goal of this project is to develop methods for efficient identification of subsequence matches in large databases of time series. Demo applications are being developed for sign language video, query by humming, and handwritten text search. Code and data from our SIGMOD 2008 paper are available here.
Nearest-neighbor retrieval and classification
Information on how to download code and datasets for BoostMap and Distance-Based Hashing can be found by clicking here.
Detecting shape classes of variable structure in cluttered images
Hand pose estimation
Code and data for this project can be downloaded as a zipped file, with size approximately 100MB, from this link.