Towards High-Quality High-Resolution Image and Video Matting

PhD Qualifying Examination


Title: "Towards High-Quality High-Resolution Image and Video Matting"

by

Miss Yanan SUN


Abstract:

Matting has long been a primary technique for image/video editing. Traditional 
matting methods outlined matting problem and made a preliminary exploration but 
their performance is limited by the low image feature representation ability. 
This issue has been addressed to a considerable extent with the introduction of 
deep neural networks into matting tasks. However, with the popularity of new 
social media (Instagram, Tik ToK, etc.) and also the advent of 4K display 
devices, higher demands are put forward on matting accuracy and the application 
scenarios. In this survey, we give a comprehensive review of recent 
high-quality high-resolution image/video matting methods. We first introduce 
how to advance general image matting performance with the help of some 
techniques such as alpha semantics. We then introduce how to promote temporal 
consistency when adapting image matting to video matting. Next,  we present how 
to achieve high human matting performance in effective and efficient manner 
when dealing with ultra high resolution images under limited computing 
resources. Their limitations are also analyzed. In the end, we conclude this 
survey by summarizing several future research directions.


Date:			Friday, 13 August 2021

Time:                  	3:00pm - 5:00pm

Zoom meeting:
https://hkust.zoom.com.cn/j/92654786122?pwd=anF0OHRwOWtGc1paRHppV1dvcDM0QT09

Committee Members:	Prof. Chi-Keung Tang (Supervisor)
 			Dr. Qifeng Chen (Chairperson)
 			Dr. Hao Chen
 			Prof. Yu-Wing Tai


**** ALL are Welcome ****