SouYoung Jin

University of Massachusetts at Amherst

PhD Candidate

SouYoung Jin is a Ph.D. Candidate in the College of Information and Computer Sciences, University of Massachusetts, Amherst, working with Prof. Erik Learned-Miller in the Computer Vision Lab. Her research interests are in Computer Vision and Machine Learning. SouYoung was a research intern at Microsoft Research AI in Redmond under the supervision from Dr. Lei Zhang. She received her master’s degree from KAIST and bachelor’s degree from Dongguk University with the honor of the first in her class. She is actively reviewing top conference/journal papers.

Research Abstract:

I focus my research in the area of computer vision and machine learning. Specifically, I am very interested in topics related to video understanding. My ultimate research goal is to build a comprehensive AI system that can understand human behaviors and closely interact with humans such as Jarvis in the movie Iron Man. This involves building the system with object/character detection/recognition/tracking, event/action/scene recognition, common-sense reasoning, and sentiment analysis. Enormous amount of videos publicly available on the internet are great resources to train the AI system. However, a very few of these online available videos have been labeled, and annotating videos for supervised training is very expensive. During my PhD studies, I explored how to automatically mine meaningful information from large-scale videos and presented a fully automatic approach to group faces in videos without any supervision (Jin et al., ICCV17). I also proposed to expand training examples by automatically mining hard examples from videos to improve the performance of the existing object detectors (Jin et al., ECCV18).