Abstract. This paper addresses a task of variable selection which consists in choosing a subset of variables that is sufficient to predict the target label well. Here instead of tr...
Aging has considerable visual effects on the human face and is difficult to simulate using a universally-applicable global model. In this paper, we focus on the hypothesis that th...
Recently, several real-time soft shadow algorithms have been introduced which all compute a single shadow map and use its texels to obtain a discrete scene representation. The res...
This paper proposes a novel method for blindly separating unobservable independent component (IC) signals based on the use of a genetic algorithm. It is intended for its applicati...
We propose a highly efficient framework for penalized likelihood kernel methods applied to multiclass models with a large, structured set of classes. As opposed to many previous a...