A New Hybrid Approach for Efficient Emotion Recognition using Deep Learning

Facial emotion recognition has been very popular area for researchers in last few decades and it is found to be challenging complex task due large intra-class changes. Existing frameworks this type of problem depends mostly on techniques like Gabor filters, principle component analysis (PCA), independent analysis(ICA) followed by some classification trained given videos images. Most these works significantly well image database acquired limited conditions but not perform with the dynamic images having varying faces In past years, various researches have introduced framework facial using deep learning methods. Although they work well, there always gap their research. research, we hybrid approach based RNN CNN which are able retrieve important parts achieve good results EMOTIC, FER-13 FERG. We also show that our accomplish promising accuracies datasets.