Video segmentation through multiscale texture analysis

Abstract

Segmenting a video sequence into different coherent scenes requires analyzing those aspects which allow finding the changes where a transition is to be found. Textures are an important feature when we try to identify or classify elements in a scene and, therefore, can be very helpful to find those frames where there is a transition. Furthermore, analyzing the textures in a given environment at different scales provides more information than considering the features which can be extracted from a single one. A standard multiscale texture analysis would require an adjustment of the scales in the comparison of the textures. However, when analyzing video sequences, this process can be simplified by assuming that the frames have been acquired at the same resolution. In this paper, we present a multiscale approach for segmenting video scenes by comparing the textures which are present in their frames.