Humans are emotional beings, and their feelings influence how they perform and interact with computers. One of the most expressive modalities for humans is body posture and movement, which researchers have recently started exploiting for emotion recognition. This survey describes emerging techniques and modalities related to emotion recognition based on body movement, as well as recent advances in automatic emotion recognition. It also describes application areas and notation systems and explains the importance of movement segmentation. It then discusses unsolved problems and provides promising directions for future research. The Web extra (a PDF file) contains tables with additional information related to the article.
Download full-text PDF |
Source |
---|---|
http://dx.doi.org/10.1109/MCG.2014.106 | DOI Listing |
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!