Video completion method for static object removal based on GLCM

Recovering and restoring the missing information after static video object removal is an active area in video processing. Reconstructions of damaged video parts are the main objectives of video completion methods. One of the main problems of most related video completion and video inpainting techniques is the processing time. They consume more time in extensive search for finding the best patch to restore areas with defects. If the affected area is wide, visual artifacts may appear. A video completion method is presented in this paper to recover static damaged area without the extensive patch search step. The proposed method uses Gray Level Co-occurrence Matrix (GLCM) and a filling method from 3 directions. The experimental results showed the good performance of the proposed method in handling the spatial and temporal consistency. Moreover, compared to existing methods, the processing time is reduced from tens of minutes to hundreds of milliseconds. © 2015 IEEE.