EXPRESSION RECOGNITION SYSTEM BASED ON DEEP LEARNING FRAMEWORK
Keywords:
Facial expression recognition, convolutional neural network, feature extraction, Tenserflow framework.Abstract
Due to the changes of expression , background , position and noise , the automatic recognition of facial expression image is a challenge for computer. The system uses the face detection module in OpenCV and Dlib library , loads 68 key point detection models to detect faces , and annotates the key points on the image. The Fer2013 database is trained to get the position information of 68 key points on the face. The expression set is predicted by the classifier , and the predicted probability is displayed visually.References
[1] Bo , C. , et al. "The Hierarchical Beta Process for Convolutional Factor Analysis and Deep Learning." ICML Omnipress , 2020.
[2] Memon , F. A. , et al. "Predicting Actions in Videos and Action-Based Segmentation Using Deep Learning." IEEE Access PP.99 ( 2021 ) :1-1.
[3] Shahid , A. R. , S. Khan , and H. Yan . "Contour and region harmonic features for sub-local facial expression recognition." Journal of Visual Communication and Image Representation 73.2 ( 2020 ) :102949.
[4] Zhang , F. , et al. "A Unified Deep Model for Joint Facial Expression Recognition , Face Synthesis , and Face Alignment." IEEE Transactions on Image Processing 29 ( 2020 ) :6574-6589.
[5] Trimech , I. H. , A. Maalej , and N. Amara . "Facial Expression Recognition Using 3D Points Aware Deep Neural Network." Traitement du Signal 38.2 ( 2021 ) :321-330.
[6] Wang X , Huang J , Zhu J , Yang M , Yang F. "Facial expression recognition with deep learning. " Proceedings of the 10th International Conference on Internet Multimedia Computing and Service. New York , NY , USA: Association for Computing Machinery , 2018