Tuesday, November 13, 2012

Paper Blogs 08



Reference Paper (From Maryam's blog post)

HEFES: An Hybrid Engine for Facial Expressions Synthesis to control human-like androids and avatars

Daniele Mazzei, Nicole Lazzeri, David Hanson and Danilo De Rossi 



Overview of the Paper

Facial expression analysis is very important in the field of affective computing. Affective computing is the section of computer science which deals with detecting and expressing human emotions and different affective states. The Facial Action Coding System (FACS) is a guide to study the facial expressions of human. This method is based of anatomical properties of human face. This model is used to synthesize human expression in avatar or robotic faces. FACS works with different action units. Action units refer to a single or combination of muscles. The authors tested their proposed method on both physical humanoid robot and also in 3-D virtual avatar.



Evaluation and Validity of the Paper



Human-like robots are used widely now-a-days in health sectors. One of the examples of human-like robot usage is in an alternative therapy for children with autism disorder (ASD). ASD children are facing difficulties to communicate with human. There are some experiments in health section is to see who well they communicate with virtual agents or human-like avatars. There are several algorithms like HEFES which is used to synthesis human emotions in virtual agents face based on FACS.

In this paper, the authors designed an experiment with two groups of children to test the capability of HEFES algorithm in conveying emotion to the robot. One group of children is normal children, and the other group of children is children with ASD. Both groups of children are interacting with robots both individually and under the supervision of a therapist. The result of this study shows that happiness, anger, and sadness are recognized by both groups of children. This is because body movements and verbal cues play a great role in conveying these emotions.


Improvement Scopes


One extension of this paper may be considering other non-verbal features like eye-gaze and head movements. These two non-verbal behavior plays very important role in emotion detection, and emotion synthesis. 

Further Reading


One of the interesting articles, which are cited by this paper, is “Exploring the Aesthetic Range for Humanoid Robots”, by David Hanson [2]. In this paper, the author performed experiment on human reaction to near realistic androids.

References


[1] D.Mazzei, N.Lazzeri, D.Hanson, and D. De Rossi, HEFES: an Hybrid Engine for Facial Expressions Synthesis to control, in proceeding BIOROB 2012 proceedings, 2012.
[2] D. Hanson, “Exploring the aesthetic range for humanoid robots,” in Proceedings of the ICCS CogSci 2006 Symposium Toward Social Mechanisms of Android Science. Citeseer, 2006, p. 1620.



No comments:

Post a Comment