publication . Article . 2021

Pairwise Ranking Network for Affect Recognition

Zoumpourlis, Georgios; Patras, Ioannis;
Open Access
  • Published: 26 Jun 2021
  • Publisher: Zenodo
Abstract
In this work we study the problem of emotion recognition under the prism of preference learning. Affective datasets are typically annotated by assigning a single absolute label, i.e. a numerical value that describes the intensity of an emotional attribute, to each sample. Then, the majority of existing works on affect recognition employ sample-wise classification/regression methods to predict affective states, using those annotations. We take a different approach and use a deep network architecture that performs joint training on the tasks of classification/regression of samples and ordinal ranking between pairs of samples. By treating input samples in a pairwis...
Subjects
ACM Computing Classification System: ComputingMethodologies_PATTERNRECOGNITION
Any information missing or wrong?Report an Issue