Shaurya Srivastava

The aim of the project was to research various methods detecting and recognizing various facial expressions, and to create an application which could identify a person's facial expression through a webcam feed in real time

The following research papers were studied for the project :

• P. Viola & M.Jones, “Rapid Object Detection using a Boosted Cascade of Simple Features”, IEEE Conference on Computer Vision and Pattern Recognition. Issue 1 pp. 511-518.

• P. Burkert, F. Trier, M. Z. Afzal, A. Dengel, and M. Liwicki, “Dexpression:Deep Convolutional Neural Network for Expression Recognition”, Computing Research Repository, vol abs/1509.05371.
Available: http://arxiv.org/abs/1509.05371

• A. De, A. Saha, and Dr M.C. Pal (2015). “A Human Facial Expression Recognition Model Based on Eigen Face Approach”, Procedia Computer Science, vol 45, pp. 282-289
https://www.sciencedirect.com/science/article/pii/S1877050915003786

Based upon this research, a classification model was created using SciKit-Learn and the FER2013 dataset. This model along with OpenCV was used to sucessfully create an application which could take images from a live webcam feed, and classify them into expressions.

Currently improving the model to increase accuracy. Code coming soon!