Feature Based Method for Human Facial Emotion Detection using Optical Flow Based Analysis

Computer has been widely deployed to our daily lives, but human computer interaction still lacks intuition. Researchers intend to resolve these shortcomings by augmenting traditional systems with human like interaction mechanism. Today, dedicated hardware often infers the emotional state from human body measures. These have been a considerable amount of research done into the detection and implicit communication channels, including facial expressions. Most studies have extracted facial features for some specific emotions in specific situations. In this paper we uses a feature point tracking technique applied to five facial image regions to capture basic emotions. The used database contains 219 images, 10 Japanese female, six expressions and one neutral. We use grayscale images which are ethically not diverse. We use optical flow based analysis to detect emotions from human facial image data. Our proof of data demonstrates the feasibility of our approach and shows promising for integration into various applications.