IJCTT-International Journal of Computer Trends and Technology
Detecting people in images is key for several important application domains in computer vision. This paper presents an in-depth experimental study on pedestrian classification; multiple feature classifier combinations are examined with respect to their performance and efficiency. In investigate global versus local, as exemplified by PCA coefficients. In terms of classifiers, consider the popular Support Vector Machines (SVMs), Adaptive boost with SVM. Experiments are performed on a large data set consisting of 4,000 pedestrian and more than statistically meaningful results are obtained by analyzing performance variances caused by varying training and test sets.