A Study of Dimensionality Reduction Impact on an Approach to People Detection in Gigapixel Images

摘要: Digital images are found in several sizes and easily displayed on a computer screen using techniques that can reduce their dimensions. Moreover, algorithms used to process perform tasks, for instance, detection of people. Recently, gigapixel emerged, providing huge amount data; however, people have been usually tested only regular size images. This paper presents an impact analysis the resolution reduction People detectors were trained with INRIA CALTECH data sets results show that, although provide false positive rate, significantly decreases number bounding boxes positives, increasing rate missing