Emotion identification method using RGB information of human face

Shinya Kita, Akira Mita

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

Recently, the number of single households is drastically increased due to the growth of the aging society and the diversity of lifestyle. Therefore, the evolution of building spaces is demanded. Biofied Building we propose can help to avoid this situation. It helps interaction between the building and residents' conscious and unconscious information using robots. The unconscious information includes emotion, condition, and behavior. One of the important information is thermal comfort. We assume we can estimate it from human face. There are many researchs about face color analysis, but a few of them are conducted in real situations. In other words, the existing methods were not used with disturbance such as room lumps. In this study, Kinect was used with face-tracking. Room lumps and task lumps were used to verify that our method could be applicable to real situation. In this research, two rooms at 22 and 28 degrees C were prepared. We showed that the transition of thermal comfort by changing temperature can be observed from human face. Thus, distinction between the data of 22 and 28 degrees C condition from face color was proved to be possible.

Original languageEnglish
Title of host publicationProceedings of SPIE - The International Society for Optical Engineering
PublisherSPIE
Volume9435
ISBN (Print)9781628415384
DOIs
Publication statusPublished - 2015
EventSensors and Smart Structures Technologies for Civil, Mechanical, and Aerospace Systems 2015 - San Diego, United States
Duration: 2015 Mar 92015 Mar 12

Other

OtherSensors and Smart Structures Technologies for Civil, Mechanical, and Aerospace Systems 2015
CountryUnited States
CitySan Diego
Period15/3/915/3/12

    Fingerprint

Keywords

  • Biofied Building
  • Face color
  • ICA
  • Thermal comfort

ASJC Scopus subject areas

  • Applied Mathematics
  • Computer Science Applications
  • Electrical and Electronic Engineering
  • Electronic, Optical and Magnetic Materials
  • Condensed Matter Physics

Cite this

Kita, S., & Mita, A. (2015). Emotion identification method using RGB information of human face. In Proceedings of SPIE - The International Society for Optical Engineering (Vol. 9435). [943531] SPIE. https://doi.org/10.1117/12.2084083