Online teaching and learning has recently turned out to be the order of the day, where majority of the learners undergo courses and trainings over the new environment. Learning through these platforms have created a requirement to understand if the learner is interested or not. Detecting engagement of the learners have sought increased attention to create learner centric models that can enhance the teaching and learning experience. The learner will over a period of time in the platform, tend to expose various emotions like engaged, bored, frustrated, confused, angry and other cues that can be classified as engaged or disengaged. This paper proposes in creating a Convolutional Neural Network (CNN) and enabling it with residual connections that can enhance the learning rate of the network and improve the classification on three Indian datasets that predominantly work on classroom engagement models. The proposed network performs well due to introduction of Residual learning that carries additional learning from the previous batch of layers into the next batch, Optimized Hyper Parametric (OHP) setting, increased dimensions of images for higher data abstraction and reduction of vanishing gradient problems resulting in managing overfitting issues. The Residual network introduced, consists of a shallow depth of 50 layers which has significantly produced an accuracy of 91.3% on ISED & iSAFE data while it achieves a 93.4% accuracy on the Daisee dataset. The average accuracy achieved by the classification network is 0.825 according to Cohens Kappa measure.

No abstract

Recognizing facial features to detect emotions has always been an interesting topic for research in the field of Computer vision and cognitive emotional analysis. In this research a model to detect and classify emotions is explored, using Deep Convolutional Neural Networks (DCNN). This model intends to classify the primary emotions (Anger, Disgust, Fear, Happy, Sad, Surprise and Neutral) using progressive learning model for a Facial Expression Recognition (FER) System. The proposed model (EmoNet) is developed based on a linear growing-shrinking filter method that shows prominent extraction of robust features for learning and interprets emotional classification for an improved accuracy. EmoNet incorporates Progressive- Resizing (PR) of images to accommodate improved learning traits from emotional datasets by adding more image data for training and Validation which helped in improving the model’s accuracy by 5%. Cross validations were carried out on the model, this enabled the model to be ready for testing on new data. EmoNet results signifies improved performance with respect to accuracy, precision and recall due to the incorporation of progressive learning Framework, Tuning Hyper parameters of the network, Image Augmentation and moderating generalization and Bias on the images. These parameters are compared with the existing models of Emotional analysis with the various datasets that are prominently available for research. The Methods, Image Data and the Fine-tuned model combinedly contributed in achieving 83.6%, 78.4%, 98.1% and 99.5% on FER2013, IMFDB, CK+ and JAFFE respectively. EmoNet has worked on four different datasets and achieved an overall accuracy of 90%.

No abstract

No abstract