REAL-TIME PROCESSING TECHNIQUE FOR CONTENT-AWARE IMAGE RESIZING OF VIDEO

In this paper, a new real-time video scaling technique that preserves the dominant contents of video is proposed. Because a correlation exists between consecutive frames in video, determining the seam of the current frame with reference to the seam of the previous frame allows us to present an effective real-time video scaling technique without additional visual artifacts and analyzing the entire video. For this purpose, frames that have similar features in video are classified into a shot, and the first frame of a shot is resized using conventional seam carving on static images to preserve the important contents of the image as much as possible. At this time, the information about the seam extracted to convert the image size is saved, and the size of the next frame is controlled with reference to the seam information stored in the previous frame. And then, above process is repeated. The proposed algorithm has a fast processing speed similar to the bilinear method, while preserving the main content of an image to the greatest extent possible. In addition, because the memory usage is remarkably small compared with the existing seam carving method, the proposed algorithm is usable in mobile devices, which have many memory restrictions. Computer simulation results indicate that the proposed technique provides a better objective performance, subjective image quality, and content conservation than conventional algorithms.