Introduction

Ubiquitous computing ( ubicomp) is a post-desktop model of Human Computer Interaction (HCI) in which information processing has been thoroughly integrated into everyday objects and activities. HCI is a very recent area of thrust as the human world is getting more and more digital and one wants the digital systems to behave as close as to a human being. This requires the devices or the information processing systems to mimic the human behavior precisely. A facial expression is indeed a human behavior which reflects the facial changes in response to a person’s internal emotional states, intentions or social communications. We have already developed an Automatic Facial Expression Recognition Engine (AFERE) which is able to identify 6 basic facial expressions from image sequences (in offline mode). Work is going on at present to develop Automatic Facial Expression Recognition Engine for online mode using camera. Principal task of AFERE is to analyze the facial expressions based on muscle movements.

Facial Expression Recognition is a means of non-verbal communication for human-beings. Mehrabian indicated that the verbal part (i.e. spoken words) of a message contributes only for 7% to the effect of message as a whole, the vocal part (e.g. voice intonation) contributes 38%, while facial expression contributes 55% to the effect of spoken message. Facial expression analysis has been an active research topic for behavioral scientists since the work of Darwin in 1872. In 90’s, Automatic Facial Expression Analysis (AFEA) research was brought into focus through the pioneering work of Bruce, Takeuchi and Nagao, Hara and Kobayashi. AFEA receives the attention of the computer vision community to be considered as a potential research topic in the context of aforementioned application areas. Through the research of the psychologists, behavioral scientists it has been established that facial expressions can convey emotion, can express intentions. Cognitive processes, physical effort, or other intra or interpersonal meanings. The principal task of AFE A is to analyze the facial expressions based on visual cues available in an image either still or video.

Transfer of Technology to the Industry

Initial Offline version of AFERE has been delivered at NICT Japan to be integrated for Robotics as part of JST-DST project.

Major Users of Products and Solutions

The product is designed for psychoanalysis of mentally retarded child and field trial is going on along with professional psychologists. Once the software is standardized it will be used in medical community.

Academic Partnership

The problem is now formulated as a post-desktop model of Human Computer Interaction(HCI) in which information processing has been thoroughly integrated into the activities of patient suffering from mental disorders like autism (low IQ) providing necessary information to psychologists for analysis. For this joint academic partnership is going on with Psychology Department of University of Calcutta, Rajabazar Science College to develop software for psychotherapeutic treatment.

Click here to know about Automatic Facial Expression Recognition Engine »