Project Kenmerken Leren
De Emotional Analyzer
let's face it
Between 1500 and 1503, Leonardo Da Vinci painted his world famous Mona Lisa. It was probably the wife of silk dealer Francesco del Giocondo who modeled for this masterpiece. For centuries, people have been speculating about the enigmatic smile on her face. What was this woman thinking about while being painted? And which emotions did the painter himself put in the painting? In 2005 MultimediaN succeeded in analyzing the mysterious Mona Lisa smile by using a specially developed Emotion Analyzer. The woman was 83% happy, 9% indicates that she felt disgust, she was afraid for 6% and something was making her angry (2%). The Emotion Analyzer can use still images and video as input. So, what are your emotions?
The Emotion Analyzer reads human emotions with the help of a webcam. After a person has taken place behind the webcam, he will be asked to put on a neutral face. After the computer software has measured this neutral face, the person can show different emotions by changing his facial expressions. The software will measure the corresponding emotions and put them in percentages.
Innovation
Emotions play an important role in human interaction. Even when a person is sitting behind his computer screen, things can become pretty emotional. It is true that modern computers are able to interact with people in a more human way than before (speech recognition, recognizing sign language and the use of joysticks), but responding to human emotions was never possible before. MultimediaN helped consolidate the software to recognize facial expressions automatically and to classify these expressions into several emotions (happy, surprised, sad, afraid, angry and disgusted). The software is also capable of learning new expressions on the basis of just a few examples.
Technology
Just a few similar systems exist, which are mainly in the stage of 'proof-of-concept' and were developed in an academic environment. An example of a commercial application is the one by Logitec. Their latest webcams enable 'face tracking' and analysis of facial expressions. In contrast to this system the Emotion Analyzer by MultimediaN functions on the basis of learning to recognize concepts under varying recording circumstances, such as lighting (light source), shadows, complexion, random movements and the size of the face (scaling). Furthermore, the system is resistant to large occlusions (coverings) and surrounding people and objects. Even if someone puts his hand in front of his face, the Emotion Analyzer can do its analyzing job.
Applications
The Emotion Analyzer software can be applied by several user groups. Anybody who wants to analyze facial expressions on emotions can use the software. Unilever could measure if eating ice cream makes you happy. LogicaCMG could use it to detect aggression at soccer matches. Publicity agencies could analyze peoples' emotions while looking at a certain commercial. Other application areas could be the medical world (for example schizophrenia), games and monitoring people during interviews.
A concrete spin off is a new website developed in cooperation with Ilse Media (www.visual-recognition.nl). Visitors can upload an image of a face for free. This face will be analyzed, after which the results are shown. Another concrete application is Pong, a ping pong game steered by emotions. Science center NEMO in Amsterdam will take up this game as a permanent exhibit for its visitors. In the construction of Pong NEMO cooperates with MultimediaN amongst others.
Partners
The Emotional Analyzer (N1)
Video Search Engine (N1)
Affective Mirror (N2)
Excercise in Immersion 4 (N2)
StreetTivo (N3 and N5)
The Investigator's Dashboard (N6)
The Surveillance Dashboard (N6)
On The Move (N6 and N9MI)
Concert Video Browser (N7 and N9MI)
Cultural Search Engine (N9C)