Fuzzy qualitative approach for micro-expression recognition

Chern Hong Lim, Kam Meng Goh

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

    4 Citations (Scopus)


    Micro-expression recognition has received increasing attention in the field of computer vision nowadays. Many state-of-the-art approaches have been reported but it can be seen that most of the results are capped at a certain level of accuracy. This is due to the ambiguity that abounded during the extraction of extremely short period of facial movements. These ambiguities deteriorate the performance of the overall recognition rate if using crisp classifier. This paper proposed to study the micro-expression as a non-mutual exclusive classification problem and examine the effectiveness of multi-label classification in micro-expression recognition by using the Fuzzy Qualitative Rank Classifier (FQRC). In addition, the extension of FQRC with feature selection and part-based model is proposed which shows promising results after tested on CASME II dataset.

    Original languageEnglish
    Title of host publicationProceedings - Ninth Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2017
    EditorsChang-Su Kim, Wai Lam Hoo
    Place of PublicationPiscataway NJ USA
    PublisherIEEE, Institute of Electrical and Electronics Engineers
    Number of pages6
    ISBN (Electronic)9781538615423, 9781538615430
    ISBN (Print)9781538615430
    Publication statusPublished - 2017
    EventAnnual Summit and Conference of the Asia-Pacific-Signal-and-Information-Processing-Association (APSIPA) 2017 - Kuala Lumpur, Malaysia
    Duration: 12 Dec 201715 Dec 2017
    Conference number: 9th
    https://ieeexplore.ieee.org/xpl/conhome/8270695/proceeding (Proceedings)


    ConferenceAnnual Summit and Conference of the Asia-Pacific-Signal-and-Information-Processing-Association (APSIPA) 2017
    Abbreviated titleAPSIPA ASC 2017
    CityKuala Lumpur
    Internet address

    Cite this