The face is the most signi_cant part of the human body for our daily mutual interaction. We look at the eyes of the person we are speaking to in order to learn whether the message contained in our words is being grasped or not. We rise our eyebrows, and open our eyes and our mouth wide to express surprise, while frowning shows anger. Not only face gestures encode our state of mind, but do the physical characteristics of our face identify us as unique individuals in a very crowded world. Since the _rst anthropometric works by Bertillon [Bert1881] and Galton [Galt1888] in the 19th century, the face has been used to register the between-class variations among di_erent people in a group, as well as the inherent or within-class variations of the face of a single subject. This is to say that we measure what distinguishes us from the rest of the people, as well as how wide the range of our iometric spectrum" is. This information can then be used to recognize the identity of people, as long as their facial features have been previously collected. Of course this process had to be done by hand. Fortunately, with the advent of potent, fast, and a_ordable computers in the last two decades, these face recognition techniques have gone automatic. Due to the recent increasing necessity of systems providing assistance for security reinforcement in crowded environments, such as airports or train stations, this kind of works are receiving more and more attention in a global scale. The original work presented in this document corresponds to this category...
The face is the most signi_cant part of the human body for our daily mutual interaction. We look at the eyes of the person we are speaking to in order to learn whether the message contained in our words is being grasped or not. We rise our eyebrows, and open our eyes and our mouth wide to express surprise, while frowning shows anger. Not only face gestures encode our state of mind, but do the physical characteristics of our face identify us as unique individuals in a very crowded world. Since the _rst anthropometric works by Bertillon [Bert1881] and Galton [Galt1888] in the 19th century, the face has been used to register the between-class variations among di_erent people in a group, as well as the inherent or within-class variations of the face of a single subject. This is to say that we measure what distinguishes us from the rest of the people, as well as how wide the range of our iometric spectrum" is. This information can then be used to recognize the identity of people, as long as their facial features have been previously collected. Of course this process had to be done by hand. Fortunately, with the advent of potent, fast, and a_ordable computers in the last two decades, these face recognition techniques have gone automatic. Due to the recent increasing necessity of systems providing assistance for security reinforcement in crowded environments, such as airports or train stations, this kind of works are receiving more and more attention in a global scale. The original work presented in this document corresponds to this category...