A survey on automatic multimodal emotion recognition in the wild

Garima Sharma, Abhinav Dhall

Research output: Chapter in Book/Report/Conference proceedingChapterpeer-review

33 Citations (Scopus)

Abstract

Affective computing has been an active area of research for the past two decades. One of the major component of affective computing is automatic emotion recognition. This chapter gives a detailed overview of different emotion recognition techniques and the predominantly used signal modalities. The discussion starts with the different emotion representations and their limitations. Given that affective computing is a data-driven research area, a thorough comparison of standard emotion labelled databases is presented. Based on the source of the data, feature extraction and analysis techniques are presented for emotion recognition. Further, applications of automatic emotion recognition are discussed along with current and important issues such as privacy and fairness.

Original languageEnglish
Title of host publicationIntelligent Systems Reference Library
Place of PublicationCham, Switzerland
PublisherSpringer
Chapter3
Pages35-64
Number of pages30
ISBN (Electronic)978-3-030-51870-7
ISBN (Print)978-3-030-51869-1
DOIs
Publication statusPublished - 2021
Externally publishedYes

Publication series

NameIntelligent Systems Reference Library
Volume189
ISSN (Print)1868-4394
ISSN (Electronic)1868-4408

Fingerprint

Dive into the research topics of 'A survey on automatic multimodal emotion recognition in the wild'. Together they form a unique fingerprint.

Cite this