A feature based approach for multiple temporal texture detection

In this paper we address the problem of detecting the presence of multiple temporal textures in an image sequence. Temporal textures are characterized by their motion pattern based features and we thus attempt an approach to detect the presence of multiple temporal textures in an image sequence from the feature space. We have established the correspondence between first order motion features of temporal textures and that of their mixture in an image sequence. We have developed a Feature based Multi-Texture Detection (FMTD) algorithm and empirically established its correctness by conducting detection experiments with a diverse set of temporal texture mixtures achieving 92.5% average accurate detection rate.