post by Siyang Song (2016 cohort)
At the end of 2019, as a Horizon CDT student at Nottingham University, I was attending a workshop (called Computer Vision for Physiological Measurement) in Seoul, South Korea. This workshop mainly focused on the research of applying recent advances in computer vision to measure the human physiological status.
This year, there were more than 50 people from company or academic institutions attending this workshop and 19 of us were giving talks to share the research we did and discussing the potential future research directions.
During this workshop, I gave a talk about how to apply state-of-the-art machine learning techniques to automatically detect emotions from people’s faces. In particular, it utilised people’s facial muscle movements to infer emotional status. This technique can be further applied to other purposes, as facial dynamics can reflect many different human statuses.
More importantly, how to apply such techniques to benefit our daily life was also discussed. For example, it can be further extended to make a quick and objective judgment about someone’s mental health, such as depression, or predict someone’s personality. Specifically, fast and automatically understanding human personality is important in employment. It can help employers to better recognise which candidates are more suitable for the job, and more willing to work in a group. Mental healthcare is another potential application. For example, while it is expensive and time-consuming to find mental health experts to diagnose mental health, such a technique can provide a cheap, quick, and objective assessment to most patients as well as provide more useful information for related doctors.
In short, such techniques have great potential to improve the business and quality of our life. For investors, it could be a good direction to invest money and time.
Since this workshop was with ICCV conference, at the end of this event, I had a great time in the banquet and had nice chats with other attendees.