Christian Moraitis, a student in the B.S. in Informatics program, has won the “Best Paper Award” for the research he conducted for his capstone project and which he presented at this year’s Hellenic American University Student Research Conference on May 13, 2023. During his presentation, the undergraduate Informatics student discussed his work in developing a tool that could eventually enable deaf and hard-of-hearing persons to use their native language to communicate online with English-language speakers who do not know ASL.
In his research, Christian used machine learning and AI-enabled object detection to create a tool that would read hand gestures in real-time, predict the corresponding American Sign Language (ASL) signs, and then compile them into a sentence. The translated caption text could then be converted into speech.
Given the constraints of time and resources, he focused on having the object detection algorithm predict the letters of the American Sign Language (ASL) alphabet. The algorithm was trained on a dataset of over 3,000 photos of signed (gestured) letters that he created himself.
The conference organizer and Director of the Research Institute, Dr. Panagiotis Kalozoumis, called the project a remarkable outcome, noting that “Christian adeptly applied the knowledge acquired from the program and combined it with his exceptional creativity.” Dr. Kalozoumis, who is also Assistant Professor and Director of Informatics Programs, added that the young software engineer’s work aptly “reflects his hard work and dedication and the rigorous standards of the BSI program.”
The idea for the project came to Christian during the COVID lockdown and the boom in videoconferencing and virtual meetings collaboration that the containment measures triggered. Christian says that while the digital platforms provided a way to keep collaborating and communicating during the quarantine for many, that was less true for deaf and hard-of-hearing persons.
Even though the project was a prototype that detected and identified letters as they were being signed, the technology, given sufficient resources, could eventually be scaled to identify and predict signs as well. Still, Christian found the project very satisfying. He says that the capstone was an ideal opportunity to pursue an idea that had been on his mind since the lockdown. “I wanted to use it to do something I couldn’t otherwise do and at the same time to learn new things and to make something of value.”
This process of taking an idea from conception to implementation is one of the things that attracted him to the field in the first place. “I was always opening up my toys when I was a kid to see how they worked. And then in high school, playing video games, I would wind up thinking, how did they put this together? I think that really sparked my interest in IT and especially development.”
As a high school student, he knew what he wanted to study and where. Initially he was hoping to study in the US—his mother is from New Jersey—but that wasn’t possible financially. His next steps, he said, was to look at American university programs in Greece. “Hellenic American University seemed to be a good fit for what I wanted to do. And it has been. I’ve fully enjoyed being a student here.”
Christian graduated this June. He managed not only to finish his B.S. in Informatics in just under four years, but to do so while working a series of jobs, including a stint as factory worker and a position managing a surfing and watersport center in Rhodes and later as a software engineer for Infralabs Ltd. He’s now working in the Netherlands as a full stack developer for a firm called Pluxbox, where he creates software tools for the company’s platform.
The capstone project he did at Hellenic American University, however, left its trace in his plans for the near future. He hopes to pursue graduate studies in machine learning and computer vision.
Image: Dr. Panayotis Kalozoumis, Christian Moraitis (BSI ’23), and Dr. Themis Kaniklidou.
