Jinwoo Choi will be heading to Snowmass Village, Colorado, in March to present “Unsupervised and Semi-Supervised Domain Adaptation for Action Recognition from Drones” during the 2020 Winter Conference on Applications of Computer Vision. WACV is a premier meeting of the IEEE Computer Society Technical Committee on Pattern Analysis and Machine Intelligence.
Choi, a Ph.D. student in the Bradley Department of Electrical and Computer Engineering, will give both a short oral presentation and a poster presentation on the paper.
Last month, he was in Vancouver, Canada, to present a poster, “Why Can’t I Dance in the Mall? Learning to Mitigate Scene Bias in Action Recognition,” at NeurIPS 2019.
“I was attracted to Virginia Tech and DAC by the students and faculty,” Choi said. “Being surrounded by people doing good research in data science gives me inspiration and keeps me motivated.”
Choi said his interest in computer vision and machine learning lies more specifically in making machines to understand what is going on in a video.
“Video understanding is a relatively under-explored area in the computer vision community,” he said, “but it can be crucial. For example, autonomous driving vehicles need to tell what pedestrians around the vehicles are doing in order to prevent accidents.”
Choi’s internship last summer was in the Media Analytics Department at NEC Labs America in San Jose, California, where he conducted research on unsupervised domain adaptation for videos. He has submitted work from this internship experience for consideration to an upcoming conference.
His projected graduation date is December 2020.
Choi earned a bachelor’s degree in electrical engineering and a master’s degree in electrical engineering and computer science from Seoul National University, Korea.
Eventually, he said, he would like to return to South Korea as a professor to continue his own research on computer vision/machine learning and teach students how to do research.