Title: Intelligent method of texture reconstruction in video sequences based on neural networks

Authors: Margarita N. Favorskaya; Mikhail V. Damov; Alexander G. Zotin

Addresses: Department of Informatics and Computer Techniques, Siberian State Aerospace University, Krasnoyarsky Rabochy av., 31, Krasnoyarsk, 660014, Russia ' Department of Informatics and Computer Techniques, Siberian State Aerospace University, Krasnoyarsky Rabochy av., 31, Krasnoyarsk, 660014, Russia ' Department of Informatics and Computer Techniques, Siberian State Aerospace University, Krasnoyarsky Rabochy av., 31, Krasnoyarsk, 660014, Russia

Abstract: We have proposed a novel intelligent method of texture reconstruction which is based on texture estimations of smoothness, structural performance, and isotropy into a missing data region using separated neural networks. Three strategies of boundaries' interpolation of various texture regions by a fast wave algorithm are suggested. The choice of texture reconstruction method is also based on neural network approach by using spatio-temporal parameters of video sequence. The proposed technique was tested for reconstruction of subtitles, damages of information medium, and small objects with area less 8% to 12% of frame area in animation and movies. A background complexity and motions in scenes determine significantly the reconstruction results.

Keywords: texture reconstruction; neural networks; missing data; video sequences; wave algorithm; smoothness; structural performance; isotropy; animation; movies.

DOI: 10.1504/IJRIS.2013.058767

International Journal of Reasoning-based Intelligent Systems, 2013 Vol.5 No.4, pp.223 - 236

Available online: 18 Jan 2014 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article