K. Yogitha Sri Lakshmi, K.Subrahmanyam, D. Harika
Abstract: Habitual face uncovering of characters in movies has drawn significant research interests and led to many put in the ground applications. It is a difficult problem due to the huge dissimilarity in the external of each character. Although existing methods display promising results in clean environment, the performance are limited in simple movie scenes due to the noises generated during the face tracking and face clustering process. We reflect on the problem of automatically labeling appearances of characters in TV or film material with their names we present two schemes of global face-name matching based framework for robust character identification. The donations of this work include the following. 1) A noise lacking feeling quality relationship representation is incorporated. 2) We commence an edit operation based graph matching algorithm. 3) Multi faceted character changes are handled by parallel graph partition and graph matching. 4) Further than existing character identification approach, we further perform an in-depth sensitivity analysis by introducing two types of simulated noises .Finally; we give a summary of the research results.
Keywords: Face detection, person classification, video indexing