The purpose of our visual emotion recognition system is to accurately discriminate six universal emotional states namely anger, disgust, fear, happiness, sadness, surprise. Emotion recognition in humans using visual information has proven to be quite challenging and an active area of research. To the best of our knowledge, simple and extended center symmetric local binary patterns (LBP) have not been used in literature on an emotional data-set and hence we have used it in our work to get a discriminating set of features. Various state of the art classifiers like Support Vector Machines (SVM), Random Forest (RF) and Logistic Regression (LR), are used in our paper to check the performance of our model. A well-known emotional video database known as eNTERFACE’05, has been employed in our research and the highest accuracy obtained from our approach is 97.9%. Our system has yielded higher accuracy on this dataset when compared with other well-known visual emotion recognition systems in literature.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Emotion Recognition in Video Clips Using Simple and Extended Center Symmetric LBP


    Contributors:


    Publication date :

    2019-05-01


    Size :

    215968 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Drivers’ Emotions’ Recognition Using Facial Expression from Live Video Clips in Autonomous Vehicles

    Rameshtrh, Tumaati / Sanampudi, Anusha / Srijayanthis, S. et al. | Wiley | 2024


    Publishing VIDEO CLIPS in Acta Astronautica

    Gerzer, R. | Online Contents | 2011


    CLIPS++: Embedding CLIPS into C++

    United States / National Aeronautics and Space Administration | British Library Conference Proceedings | 1995


    CLIPS

    IWAHARA TOSHIO / KATO MAKOTO / HIRAKAWA KATSUYA | European Patent Office | 2017

    Free access

    DAI-CLIPS: Distributed, Asynchronous, Interacting CLIPS

    United States / National Aeronautics and Space Administration | British Library Conference Proceedings | 1995