Joy Anne Cotoner and Nathaniel Rios. 4 0
Facial expression recognition: an enhancement of the hidden markow models / 6 6 Joy Anne Cotoner and Nathaniel Rios. - - - 42 pp. 28 cm. - - - - - . - . - 0 . - . - 0 .
Thesis: (BSCS major in Computer Science) - Pamantasan ng Lungsod ng Maynila, 2005.
5
ABSTRACT: What is it about a face that tells us whether a person is happy, sad, or angry? While most of us have not thought about these factors, the same cannot be said about the many researchers who have developed the field of emotion recognition in facial expressions and its three perspectives. In recent years, considerable progress in technology has been made and that computers now can outperform humans in many face recognition tasks through the use and still development of different techniques. Just as human's capability of recognizing and distinguishing different faces, computers are now catching up. In the early 1990's the engineering community started to use these results to construct automatic methods of recognizing emotions from facial expressions in images or videos. Work on recognition of emotions from voice and videos has been recently suggested and shown to work by some experts. The logic behind using all of the temporal information is that any emotion being displayed has a unique temporal pattern. Human-computer intelligent interaction is an emerging of science aimed at providing natural ways for humans to use computers as aids. It is argued that for the computer to be able to interact with humans, it needs to have the communication skills of humans. One of these skills is the ability to understand the emotional state of the person. The most expressive way humans display emotions is through facial expressions. This work focuses on automatic facial expression recognition from the live video input using temporal cues. Methods for using temporal information have been extensively explored for speech recognition applications. Among these methods is template matching using dynamic programming methods and hidden markov models (HMM). This work exploits existing methods and proposes a new architecture of HHM for automatically segmenting and recognizing human facial expression from videos sequences. To come up a solution to this dilemma, lot of approaches used and HHM is agreed to be more promising one. The modification of HMM takes the burden of releasing from representative approximations hoping for better results. Facial expression recognition is the process by which a computer will be identify the closest emotional state of the person (www.ri.cmu.edu/pubs/pub_468.html). Selecting the person's model and the facial expression that maximizes the like hood of the test images carries out face recognition. Then, in a person dependent context expression recognition is achieved by selecting the facial expressions with maximum like hood. The most techniques to be used to determined the facial expression of the person is thru Hidden Markov Models Algorithm (http://www.cs.cmu.edu/~jjlien/Thesis/thesis2.html). The approach used for facial expression recognition is based on the Facial to extract facial expression information: (1) facial feature of the face and etc. Recognition of emotional expression in faces is a complex problem with no simple answer. While it may appear that understanding emotions is an instinct from the beginning of life, there are many individuals without this ability. Discoveries of specific structures in the brain shed new light on what triggers this skill. Finally, there are components of the face that draw our focus when understanding the emotion present, even if we are not consciously aware of them. Since the display of a certain facial expression in video is represented by a temporal sequence of facial emotions it is natural to model each expression using an HMM trained for that particular type of expression. There will be six such HMMs, one for each expression: happy (1), angry (2), surprise (3), disgust (4), fear (5), sad (6). There are several choice of model structure that can be used; the two main models are left-to-right model. In the left-to-right model, the probability of going back to the previous state is set to zero, and therefore the model will always start and end up in an existing safe. In the ergodic model every state can be reached from any other state in a finite number of time steps. The advantage of using this kind of model is that, it lies in the fact that it seems natural to model a sequential event with a model that also starts from a fixed starting state and always reaches an end state. It also involves fewer parameters and therefore is easier to train. However, it reduces the degrees of freedom the model has totry to account for the observation sequence. On the other hand, using the ergodic HMM allows more freedom for the model to account for the observation sequences and intact, for an infinite amount of training data it can be shown that the ergodic model will reduce to the left-to-right model. In this framework, face models jointly capture information about facial appearance and expression patterns so that recognition of faces and facial expressions carried at the same time. Face and facial expression recognition cooperate so that the similarity measure used for face recognition benefits from facial expression modelling. Although it is considered the most successful model used in facial expression recognition, it also has its drawback, which needs some improvements in order to provide a better results in its applications.
5
2 = =
2
2 --0------
6 --0-- 2 --------
0 2 --
--20------
--------20--
--------20--
----2
/ 2
/ 2
/
/
Facial expression recognition: an enhancement of the hidden markow models / 6 6 Joy Anne Cotoner and Nathaniel Rios. - - - 42 pp. 28 cm. - - - - - . - . - 0 . - . - 0 .
Thesis: (BSCS major in Computer Science) - Pamantasan ng Lungsod ng Maynila, 2005.
5
ABSTRACT: What is it about a face that tells us whether a person is happy, sad, or angry? While most of us have not thought about these factors, the same cannot be said about the many researchers who have developed the field of emotion recognition in facial expressions and its three perspectives. In recent years, considerable progress in technology has been made and that computers now can outperform humans in many face recognition tasks through the use and still development of different techniques. Just as human's capability of recognizing and distinguishing different faces, computers are now catching up. In the early 1990's the engineering community started to use these results to construct automatic methods of recognizing emotions from facial expressions in images or videos. Work on recognition of emotions from voice and videos has been recently suggested and shown to work by some experts. The logic behind using all of the temporal information is that any emotion being displayed has a unique temporal pattern. Human-computer intelligent interaction is an emerging of science aimed at providing natural ways for humans to use computers as aids. It is argued that for the computer to be able to interact with humans, it needs to have the communication skills of humans. One of these skills is the ability to understand the emotional state of the person. The most expressive way humans display emotions is through facial expressions. This work focuses on automatic facial expression recognition from the live video input using temporal cues. Methods for using temporal information have been extensively explored for speech recognition applications. Among these methods is template matching using dynamic programming methods and hidden markov models (HMM). This work exploits existing methods and proposes a new architecture of HHM for automatically segmenting and recognizing human facial expression from videos sequences. To come up a solution to this dilemma, lot of approaches used and HHM is agreed to be more promising one. The modification of HMM takes the burden of releasing from representative approximations hoping for better results. Facial expression recognition is the process by which a computer will be identify the closest emotional state of the person (www.ri.cmu.edu/pubs/pub_468.html). Selecting the person's model and the facial expression that maximizes the like hood of the test images carries out face recognition. Then, in a person dependent context expression recognition is achieved by selecting the facial expressions with maximum like hood. The most techniques to be used to determined the facial expression of the person is thru Hidden Markov Models Algorithm (http://www.cs.cmu.edu/~jjlien/Thesis/thesis2.html). The approach used for facial expression recognition is based on the Facial to extract facial expression information: (1) facial feature of the face and etc. Recognition of emotional expression in faces is a complex problem with no simple answer. While it may appear that understanding emotions is an instinct from the beginning of life, there are many individuals without this ability. Discoveries of specific structures in the brain shed new light on what triggers this skill. Finally, there are components of the face that draw our focus when understanding the emotion present, even if we are not consciously aware of them. Since the display of a certain facial expression in video is represented by a temporal sequence of facial emotions it is natural to model each expression using an HMM trained for that particular type of expression. There will be six such HMMs, one for each expression: happy (1), angry (2), surprise (3), disgust (4), fear (5), sad (6). There are several choice of model structure that can be used; the two main models are left-to-right model. In the left-to-right model, the probability of going back to the previous state is set to zero, and therefore the model will always start and end up in an existing safe. In the ergodic model every state can be reached from any other state in a finite number of time steps. The advantage of using this kind of model is that, it lies in the fact that it seems natural to model a sequential event with a model that also starts from a fixed starting state and always reaches an end state. It also involves fewer parameters and therefore is easier to train. However, it reduces the degrees of freedom the model has totry to account for the observation sequence. On the other hand, using the ergodic HMM allows more freedom for the model to account for the observation sequences and intact, for an infinite amount of training data it can be shown that the ergodic model will reduce to the left-to-right model. In this framework, face models jointly capture information about facial appearance and expression patterns so that recognition of faces and facial expressions carried at the same time. Face and facial expression recognition cooperate so that the similarity measure used for face recognition benefits from facial expression modelling. Although it is considered the most successful model used in facial expression recognition, it also has its drawback, which needs some improvements in order to provide a better results in its applications.
5
2 = =
2
2 --0------
6 --0-- 2 --------
0 2 --
--20------
--------20--
--------20--
----2
/ 2
/ 2
/
/