In this paper, a novel method for tracking multiple soccer players based on a support vector machine (SVM) and a particle filter is proposed for a monocular fixed soccer video. This paper considers three main topics. First, the state of a particle filter presented in a 3D world coordinate space, instead of a 2D image coordinate space, provides a more realistic model tracking because depth consi...

A probabilistic interpretation for the output obtained from a tri-class Support Vector Machine into a multi-classification problem is presented in this paper. Probabilistic outputs are defined when solving a multi-class problem by using an ensemble architecture with tri-class learning machines working in parallel. This architecture enables the definition of an ‘interpretation’ mapping which wor...

Many adaptive tracking-by-detection methods have been proposed to track object with slowly changing appearance. However, most of those methods are designed for single object tracking. This paper proposes a method for adaptively tracking multiple objects based on a modified structured Support Vector Machine (SVM). The method utilizes the inter-object constraints and the layout information, which...

This technical report provides an overview on the support vector machines that have been designed during the recent years for the identification of microRNA sequences. Our aim is to arrive in the end to identify potential ways in which the quality of such machine learning methods could be further improved, knowing that the rate of false positves they produce is still too high.

Feature selection is an important issue in many research areas. There are some reasons for selecting important features such as reducing the learning time, improving the accuracy, etc. This thesis investigates the performance of combining support vector machines (SVM) and various feature selection strategies. The first part of the thesis mainly describes the existing feature selection methods a...

This paper proposes a new method for extracting impervious surface from VHR imagery. Since the impervious surface is the only class of interest (i.e. target class), the One Class Support Vector Machine (OCSVM), a recently developed statistical learning method, was used as the classifier. Rather than use samples from all classes for training in traditional multi-class classification, the method ...

In this paper, a method for secondary structure with support vector machines is presented. The system used two layers of support vector machines, with a weighted cost function to balance the uneven class memberships. Using this method, prediction accuracy reaches 71.5%, comparable to the best techniques avaliable.

The standard -norm SVM is known for its good performance in twoclass classification. In this paper, we consider the -norm SVM. We argue that the -norm SVM may have some advantage over the standard -norm SVM, especially when there are redundant noise features. We also propose an efficient algorithm that computes the whole solution path of the -norm SVM, hence facilitates adaptive selection of th...

This paper describes an adaptive multi-modal person verification system based on speech and face images. The system adapts to noise present in the speech signal by modifying the parameters of the fusion module. Linear and Support Vector Machine (SVM) based techniques of fusing the similarity measures from speech and face modes are investigated. Experimental results obtained on the Digit Databas...