Fusion of face and voice for automatic human age estimation

Xin Geng, Erqing Fang, Kate Smith-Miles

Research output: Chapter in Book/Report/Conference proceedingConference PaperResearchpeer-review


People usually estimate other people s age based on multiple aging traits, such as face, voice, gait, hair, and body build. However, almost all existing work on automatic age estimation is based on unimodal traits. This paper presents one of the first attempts toward multimodal automatic age estimation, which is based on the fusion of face and voice. The age estimators based on face and voice are all trained by the IIS-LLD algorithm, and the fusion is at the score level through weighted sum. Experimental results show that the fusion of face and voice can achieve better age estimation performance than that based on face only.
Original languageEnglish
Title of host publicationProceedings of the 3rd International Conference on Computer Design and Applications
EditorsZ X Yin
Place of PublicationSingapore
PublisherIEEE, Institute of Electrical and Electronics Engineers
Pages311 - 314
Number of pages4
ISBN (Print)9781424494897
Publication statusPublished - 2011
EventInternational Conference on Computer Design and Application (ICCDA 2011) - Shanxi Normal University, Xi'an, China
Duration: 27 May 201127 May 2011
Conference number: 3rd


ConferenceInternational Conference on Computer Design and Application (ICCDA 2011)
Abbreviated titleICCDA 2011

Cite this