Volume : III, Issue : IV, April - 2014

Intelligent Object–Based Video Inpainting Approach For Fast Video Repairing

Mr. Abhijeet A. Chincholkar, Prof. Salim A. Chavan

Abstract :

This paper approaches improved technique to speed up the object based video Inpainting for fast Video Repairing. This approaches to reduce search area by converting video into 3D patches and 3D patches into 2D slices. Here method tries to improve existing Exemplar-Based inpainting method as well as Patch Match algorithm for reducing processing time. The main target of this method is to make better priority function. After combining completed slices, sequences of virtual contours forms to retrieve the most similar postures among number of available postures. To overcome over-smoothing problem resulting from the averaging of patches it tries to give the good results in one fourth time to fill the target region. Key posture selection method and indexing method are used to curtail the complexity of posture sequence retrieval. Synthetic posture generation method is used to generate unavailable postures and to increase the number of postures in database and tries to maintain spatial consistency and temporal motion continuity of an object simultaneously.

Keywords :

Article: Download PDF   DOI : 10.36106/ijsr  

Cite This Article:

Mr.Abhijeet A.Chincholkar, Prof.Salim A.Chavan Intelligent Object-Based Video Inpainting Approach For Fast Video Repairing International Journal of Scientific Research, Vol.III, Issue. IV Apr 2014

Number of Downloads : 1024

References :