Computer Graphics & Animation

Biography

Biography: Russell Pensyl

Abstract

Facial recognition technology is a growing area of interest, where researchers are using these new application for study in psychology, marketing and product testing and other areas. There are also application where the use of facial image capture and analysis can be used to create new methods for control, mediation and integration of personalized information into web based, mobile apps and standalone system for media content interaction. Our work explores the application of facial recognition with emotion detection, to create experiences within these domains. For mobile media applications, personalized experiences can be layered personal communication. Our current software implementation can detect smiles, sadness, frowns, disgust confusion, and anger. In a mobile media environment, content on a device can be altered, to create a fun, interactive experience, which is personally responsive and intelligent. By intersecting via direct communication between peer to peer mobile apps, moods can be instantly conveyed to friends and family – when desired by the individual. This creates a more personalized social media experience. Connections can be created with varying levels of intimacy, from family members, to close friends, out to acquaintances and further to broader groups as well. This technique currently uses an pattern recognition to identify shapes within a image field using Viola and Jones Open CV Haar-like features application [1], [2],[3] and a “feret” database [4] of facial image and support vector machine (LibSVM) [3] to classify the capture of the camera view field and identify if a face exists. The system processes the detected faces using an elastic bunch graph mapping technique that is trained to determine facial expressions. These facial expressions are graphed on a sliding scale to match the distance from a target emotion graph, thus giving an approximate determination of the user’s mood.