From Facial Micro-Expression Recognition to Protective Movement Behavior Detection
- 14:00 10th December 2018 ( Michaelmas Term 2018 )CPS group, 2nd floor, Hobert Hooke Building (access required, pls contact Xiaoxuan Lu)
Talk Abstract
Understanding the affective state and emotional need of human has recently been an appealing research topic in computer science, psychology and industry. In this talk, I will show you two affective modalities, the facial micro-expression and bodily movement behavior.
Micro-expression, as a special form of facial expression, has recently caught more and more attention. Unlike the traditional macro-expression, micro-expression is transient and imperceptible and usually demonstrated when a person attempts to conceive his underlying emotion. I will describe two research works conducted with my colleagues in China on the recognition of 5 basic emotion types from micro-expressions. Those works aimed to solve the problem of non-uniform framerates of existing databases and the localized, sparse characteristic of micro-expression.
For my PhD study at UCL, I aim to automatically detect the level of pain and confidence of people having chronic lower back pain (CLBP) during their physical rehabilitation. Specifically, I will look into the protective behavior as an important affective clue, which CLBP patient would demonstrate during exercise due to the fear of potential pain or the underlying pain. The ultimate goal is to establish intelligent rehabilitation therapy for a wider CLBP population. In this talk, I will show some initial results we acquired toward a continuous detection of protective behavior, based on motion capture and sEMG data collected from real patients.
Speaker Bio
Chongyang Wang is a second-year PhD student at the UCL interaction centre under the supervision of Prof. Nadia Berthouze. He graduated from the Southwest University of China in Electronics and Information Engineering. As an undergraduate student, he has worked in a number research projects in the area of signal processing, image processing and machine learning and contributed to four peer-reviewed publications, one of which in a journal and one as first author. In 2016, he was ranked 1st at his school's comprehensive appraisal and won a national scholarship. He did a one-year research internship at the Institute of Psychology, Chinese Academy of Science, a very prestigious research institution in China. Finally, he was awarded two prestigious scholarship from UCL to carry out his PhD studies on the topic of developing new body sensing technology to support chronic pain physical rehabilitation.