Abstract
Dissimilarity measures are often used as a proxy or a handle to reason about data. This can be problematic, as the data representation is often a consequence of the capturing process or how the data is visualized, rather than a reflection of the semantics that we want to extract. Facial expressions are a subtle and essential part of human communication but they are challenging to extract from current representations. In this paper we present a method that is capable of learning semantic representations of faces in a data driven manner. Our approach uses sparse human supervision which our method grounds in the data. We provide experimental justification of our approach showing that our representation improves the performance for emotion classification.
Original language | English |
---|---|
Title of host publication | Proceedings - 13th IEEE International Conference on Automatic Face and Gesture Recognition, FG 2018 |
Publisher | Institute of Electrical and Electronics Engineers (IEEE) |
Pages | 179-186 |
Number of pages | 8 |
ISBN (Electronic) | 9781538623350 |
DOIs | |
Publication status | Published - 5 Jun 2018 |
Event | 13th IEEE International Conference on Automatic Face and Gesture Recognition, FG 2018 - Xi'an, China Duration: 15 May 2018 → 19 May 2018 |
Conference
Conference | 13th IEEE International Conference on Automatic Face and Gesture Recognition, FG 2018 |
---|---|
Country/Territory | China |
City | Xi'an |
Period | 15/05/18 → 19/05/18 |
Keywords
- Facial expressions
- Representation learning
- Variational auto encoder