A Novel Feedback Approach for Image/Video sampling Systems' Performances Improvement

As the application of sampling system has stretched into image/video compression and video scalable coding areas, many researches have been carried out on how to improve the sampling system’s performance. So far, a lot of approaches have been proposed, such as directional sampling method, and adaptive sampling filter scheme. Every design has its own trade-off between image/video quality and computation complexity. In this paper, we propose a feedback method to get significant improvements in various image/video sampling systems. By feeding back the system errors into the input side, the existing sampling systems’ performances can be improved greatly. Experimental results showed that with only 3 loops’ feedback, the improvements can be up to 3dB gain in PSNR together with the obvious subjective progress.