Venue: FG 2018, Xi'an, China https://fg2018.cse.sc.edu/
Paper Title: Human Behaviour-based Automatic Depression Analysis using Hand-crafted Statistics and Deep Learned Spectral Features
Background: Depression is a serious mental disorder that affects millions of people all over the world. It can negatively affect a person’s personal, work, school life, as well as sleeping, eating habits, general health, etc. and affects thoughts, behaviour, feelings, and sense of well-being. In extreme conditions, people even die by suicide.
Gaps: Current clinical standards for depression assessment are subjective, and requires extensive participation from experienced psychologists .
Motivation: Building an objective and fast automatic depression analysis approach. This approach is expected to based on the non-verbal automatic detected human behaviours, and able to construct a fixed-size video-level descriptor for variable-length video.
Mehtod:
Results:
Poster:
Role of paper within PhD:
This study is the basis of my PhD study. Based on this work, while we continuously developing a better model for practical use, we also looking to other temporal modeling methods.
Following activities
Building several models for not only depression analysis, but also personality traits and anxiety analysis.
Ppaer Link: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=8373825 http://eprints.nottingham.ac.uk/51476/1/human-behaviour-based%20camera%20ready.pdf
PS: Due to the limited pages, the paper is not perfect and explained in detalis. More details and experiments (including well organized code) will be available in the Journal version paper (available soon).