Abstract
Video blogs (vlogs) are a popular media form for people to present themselves. In case a vlogger would be a job candidate, vlog content can be useful for automatically assessing the candidates traits, as well as potential interviewability. Using a dataset from the CVPR ChaLearn competition, we build a model predicting Big Five personality trait scores and interviewability of vloggers, explicitly targeting explainability of the system output to humans without technical background. We use human-explainable features as input, and a linear model for the systems building blocks. Four multimodal feature representations are constructed to capture facial expression, movement, and linguistic usage. For each, PCA is used for dimensionality reduction and simple linear regression for the predictive model. Our system's accuracy lies in the middle of the quantitative competition chart, while we can trace back the reasoning behind each score and generate a qualitative analysis report per video.
Original language | English |
---|---|
Title of host publication | Proceedings - 30th IEEE Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2017 |
Publisher | IEEE |
Pages | 1664-1669 |
Number of pages | 6 |
Volume | 2017-July |
ISBN (Electronic) | 978-1-5386-0733-6 |
ISBN (Print) | 978-1-5386-0734-3 |
DOIs | |
Publication status | Published - 2017 |
Event | CVPRW 2017: 30th IEEE Conference on Computer Vision and Pattern Recognition Workshops - Honolulu,HI, United States Duration: 21 Jul 2017 → 26 Jul 2017 |
Conference
Conference | CVPRW 2017 |
---|---|
Country/Territory | United States |
City | Honolulu,HI |
Period | 21/07/17 → 26/07/17 |
Keywords
- Face
- Video recording
- Predictive models
- Feature extraction
- Gold
- principal component analysis
- Lips