Professional Documents
Culture Documents
Human Face Detection and Facial Expression Identification PDF
Human Face Detection and Facial Expression Identification PDF
EXPRESSION IDENTIFICATION
• For interactive human and computer interface (HCI) it is important that the
computer understand facial expressions of human.
• These movements are used to help one deliver the emotional state to other person.
• As per Mehrabian, the message conveys only 7% of its meaning in spoken part
and vocal part contributes 38%, but facial expressions have 55%.
• Darwin suggested that there is a link between mental states (e.g. sensed emotions,
belief in and thought) and muscle movements below the face skin.
• Regardless of gender or ethos the muscles that together produce the facial
expressions are worldwide alike.
Existing System
• Machine learning algorithms are used.
• Train the classifier.
• Each face is divide into non-overlapping patches.
• Number of images are used for training set.
Drawbacks
• So many data sets are needed in machine learning.
• Needs lots of images to train classifier.
• Time taken to divide image into patches and compare with each image patches in
training set.
• Time taken to compare test image with all images in training set.
Proposed System
• The proposed system can identify expression for both captured image and live
image
• Face detection or face localization, feature extraction and classification for image
using Haar-AdaBoost algorithm .
• Facial expression recognitions and gender identification using Convolutional
Neural Networks .
Advantages
• No use of training data.
• Reduces time complexity.
• Automatically face expression is identified with CNN classifier.
Architecture
Software Requirements And Specifications
• RAM : Min 4 GB
• IDE : Pycharm
References
[1] Mehrabian A.: Communication Without Words. Psychology Today, Issue 2, 53--
56 (1968).
[2] Ekman, P., & Friesen, W. V. (1971). Constants across cultures in the face and
emo-tion. Journal of Personality and Social Psychology, 17(2), 124-129.
[3] P. T. Yap, R. Paramesran and Seng-Huat Ong, "Image analysis by Krawtchouk
moments," in IEEE Transactions on Image Processing, vol. 12, no. 11, pp. 1367-
1377, Nov. 2003.
[4] A. Koutlas and D. I. Fotiadis, "An automatic region based methodology for
facial expression recognition," 2008 IEEE International Conference on Systems,
Man and Cybernetics, Singapore, 2008, pp. 662-666.
[5] Velusamy S., Kannan H., Balasubramanian A., Sharma A., Navathe B..: A
Method to Infer Emotions from Facial Action Units. IEEE ICASSP, 2028--2031
(2011).
THANK YOU