University of Science & Technology, Republic of Korea
Thomhert S. Siadari received his BS degree in Telecommunication Engineering from Telkom University, Indonesia, in 2011 and his M.Eng degree in IT Convergence Engineering from Kumoh National Institute of Technology, Republic of Korea, in 2013. Currently, he is pursuing PhD degree in ICT from the ETRI School, University of Science & Technology (https://ust.ac.kr) and also a student researcher at City & Transportation ICT Research Department, ETRI (https://etri.re.kr) supervised by Prof. Hyunjin Yoon. His research interests are mainly related to image and video understanding using deep learning in various applications including smart city and medical healthcare.
Tutorial Title: Deep Representation Learning for Recognizing Human and Object Interactions
Human–object interaction (HOI) detection aims to detect human and object locations and classify their interactions at the instance level (e.g.,, a person riding a bike, carrying a backpack, and throwing a Frisbee), which can be formulated as detecting a triplet (human, action, and object). This task is beneficial to many applications that require a deeper understanding of semantic scenes, such as video surveillance and visual question answering.