Non-iterative coding tree depth estimation for H.265/HEVC using neighboring block information

This paper presents a novel coding tree depth estimation method for H.265/HEVC using neighboring block information. There have been some complexity reduction methods for H.265/HEVC by estimating coding tree structure from neighboring blocks, while preserving almost the same image quality and compression rate. The coding tree structure is not estimated for all the frames, its interval is determined and updated by the difference between the target and the predicted complexities. However, the updating procedure is very sensitive to the target and the predicted complexities since it does not depend on the resolution of the image. Also the image quality is unstable until the estimation procedure is converged. In this paper, we propose a novel non-iterative estimation equation to determine frame insertion interval, based on the fact that the optimal interval of non-estimation frame is related to the resolution and the frame rate of moving image. The performance of the proposed method is evaluated in comparison with some conventional methods through computer simulation.