Metaheuristic optimization with deep learning mosel for automated video inpainting technique

Abstract

In recent times, video inpainting technique has received maximum attention among researchers which intends to fill the missing areas or gaps in a video utilizing the known pixels. Since the state of art video inpainting techniques exhibit high computation complexity and create seam in the target areas which occurs because of the variation in brightness or contrast of the patches. The video inpainting process helps in several video editing and restoration processes like unwanted object removal, scratch or damage rebuilding, and retargeting. Video inpainting remains a hot research topic which fills spatio-temporal holes with conceivable content in a video. It is a highly challenging process as the needs are two folded: produced content in the missing area should be semantically correct based on nearby content, and missing region is required to be filled in seamlessly to make sure that the actual holes are visually unnoticeable. We have formulated three research objectives to overcome the limitations identified in this research study. newline