Knowledge Technology Won the Most Popular Presentation Runner-up
23 November 2020
Knowledge Technoledge's paper "Adapting the Interplay between Personalized and Generalized Affect Recognition based on an Unsupervised Neural Framework", in IEEE Transactions on Affective Computing, has been selected as runner-up for the most popular presentation (journal track) at the ICDL-EpiRob 2020 conference.
This work introduces a framework for audio/visual emotion expression recognition that integrates generalized and personalized perception into one single continual emotional reading. It employs growing neural networks to allow the online clustering of perceived effects. The experiments demonstrate how our model adapts its interpretation of affective perception towards specific characteristics of an individual, and thus, improves the performance on arousal and valence recognition.
You can get more detailed information from the video of the presentation, the source code, and the full paper.