Apostolos Georgakis

Learn More
In this paper we proposed a visual speech recognition network based on Support Vector Machines. Each word of the dictionary is modeled by a set of temporal sequences of visemes. Each viseme is described by a support vector machine, and the temporal character of speech is modeled by integrating the support vector machines as nodes into Viterbi decoding(More)
The concept of rotation in the joint time-frequency plane can be exploited in order to generalize classical Fourier-based operations. It is known that filtering in rotated time-frequency domains can lead to significant performance advantages for certain types of signals as compared to conventional linear time invariant systems. In this correspondence, we(More)
The averaged instantaneous frequency (AIF) is proposed as an alternative method for the frequency analysis of surface electromyography (EMG) in the study of muscle fatigue during sustained, isometric muscle contractions. Results from performance analysis using experimental EMG signals demonstrate the low variability of the proposed frequency variable.(More)
The self-organizing map algorithm has been used successfully in document organization. We now propose using the same algorithm for document retrieval. Moreover, we test the performance of the self-organizing map by replacing the linear Least Mean Squares adaptation rule with the marginal median. We present two implementations of the latter variant of the(More)
A novel method for the suppression of cross terms in the time-frequency domain is introduced. First, interference is identified using a fractional Fourier transform-based technique. Then, auto-terms are detected, synthesized, and subtracted from the original signal. The process is repeated until all signal components are extracted. Finally, the Wigner(More)
The purpose of this study was to capture the lower limb kinematics before during and after ball impact of soccer kicking by examining the influence of both sampling rate and smoothing procedures. Nine male soccer players performed maximal instep kicks and the three-dimensional leg movements were captured at 1000 Hz. Angular and linear velocities and(More)
A new filtering algorithm is proposed for the accurate estimation of the second derivatives of kinematic signals with impacts. The algorithm operates in predetermined consecutive fractional Fourier transform domains and amounts to an overall linear low-pass filter with time-varying cutoff threshold, which can successfully accommodate the impact-induced(More)
The problem of lip contour detection is critical in the lipreading systems based on contour processing. The typical contour detection strategy based on image segmentation in homogeneous regions fails in the case when the mouth images available for lipreading are lowcontrast gray level images. Most of the solutions adopted require manual marking of some(More)