HUMAN POSE ESTIMATION AND SHAPE RECOVERY: A SURVEY

PhD Qualifying Examination


Title: "HUMAN POSE ESTIMATION AND SHAPE RECOVERY: A SURVEY"

by

Mr. Hao XU


Abstract:

Human pose estimation attempts to locate the spatial position of various parts 
of the human body, while shape recovery attempts to recover the observable 
surface in a more dense manner, so as to have a better understanding of the 
human pose, motion and behavior. Understanding the human body pose has a wide 
range of applications in the fields of medicine, sports, film industry, 
augmented reality and virtual reality. Although this field has achieved rapid 
development in recent years, there are still challenges such as occlusion, not 
conform to physical reality, consistency in temporal, generalization issues due 
to domain gap, etc. This paper conducts a in-depth survey, comparison between 
state of the art methods, and comprehensive summary of human body pose 
estimation and shape recovery based on visual inputs. Finally, regarding the 
challenges, this article also gives ideas and research suggestions to 
facilitate the future work in this field.


Date:			Thursday, 30 September 2021

Time:                  	4:00pm - 6:00pm

Zoom meeting:
https://hkust.zoom.com.cn/j/96590295258?pwd=TzlDWWo4MjFhdFRUODRIbi9vRU9KUT09

Committee Members:	Dr. Kai Chen (Supervisor)
 			Dr. Yangqiu Song (Chairperson)
 			Dr. Qifeng Chen
 			Dr. Dan Xu


**** ALL are Welcome ****