In this paper a novel method for view independent human movement representation and recognition, exploiting the rich information contained in multi-view videos, is proposed. The bi...
Nikolaos Gkalelis, Nikos Nikolaidis, Ioannis Pitas
Gait is a promising biometric cue which can facilitate the recognition of human beings, particularly when other biometrics are unavailable. Existing work for gait recognition, how...
Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA) techniques are important and well-developed area of image recognition and to date many linear discriminati...
"Pattern recognition techniques are concerned with the theory and algorithms of putting abstract objects, e.g., measurements made on physical objects, into categories. Typical...
We consider the supervised learning of a binary classifier from noisy observations. We use smooth boosting to linearly combine abstaining hypotheses, each of which maps a subcube...