Super resolution based video inpainting

Abstract

Video inpainting is the technique of filling the missing region or gaps in a video. In this technique, to fill up the large hole, high computational time is required. To overcome this issue, super resolution technique is used in video inpainting. In this paper, super resolution based video inpainting framework is proposed to fill the gaps in a video sequence. Iterated back projection based super resolution technique is used to super resolve the low resolution image. The inpainting method is applied on down-sampled low resolution image. A super-resolution algorithm is then used to enhance the image. The proposed approach is compared with state-of-art algorithms proposed by Soheil and Yunqiang. The results demonstrate that the proposed work is well suited for filling the gap in a frame of a video sequence in less time compared to the state-of-the-art algorithms.