Video Inpainting Using Seam Carving Method

Rana Zaki,Matheel Abdulmunim

doi:10.24237/djps.1301.173b

Abstract

Video Inpainting Using Seam Carving Method

Highlights

Inpainting historically painters to be removing flaw from photos and paintings manually, it was called
Image inpainting the concept exists very long years back and from the birth of computer vision, and researchers are looking for a way to do this process automatically [1], developed process to remove the certain areas or restoration a damaged area in a video is known.There are many algorithms and applications of image inpainting
Video structure plays very important role in the understanding of video, it is well known about structure contains measure of frame size, if it has a width of W pixels and a height of H pixels at Color Depth (CD) of 24 bits, and the rate at which frames are displayed in Frames Per Second (FPS)

Summary

Introduction

Inpainting historically painters to be removing flaw from photos and paintings manually, it was called (image inpainting). Image inpainting the concept exists very long years back and from the birth of computer vision, and researchers are looking for a way to do this process automatically [1], developed process to remove the certain areas or restoration a damaged area in a video is known (video inpainting).There are many algorithms and applications of image inpainting It can be used in cinema and photography for “restoration”, to remove effects such as scratches, dust spot from images, removal of superimposed text like dates, publicity, or subtitles, (called deterioration). Seam carving method Seam carving was implemented according to algorithm proposed by “Shai Avidan and Ariel Shamir in 2007” [5].The basic algorithm is elegant and quite simple Such as includes many applications (enlarging, shrink, cropping, and object removal, etc.). Seam carving is used to remove the pixels with “low energy” and avoid removing pixels with “high energy” [5]

Finding energy using edge detection

Computed Seam Cost

Video structure

Preprocessing of each frame

Reconstruction Process

Experimental Tests and Results

Conclusions