A High-Quality Adaptive Video Reconstruction Optimization Method Based on Compressed Sensing
Yanjun Zhang,Yongqiang He,Jingbo Zhang,Yaru Zhao,Zhihua Cui,Wensheng Zhang
DOI: https://doi.org/10.32604/cmes.2023.025832
2023-01-01
Computer Modeling in Engineering & Sciences
Abstract:The video compression sensing method based on multi hypothesis has attracted extensive attention in the research of video codec with limited resources. However, the formation of high-quality prediction blocks in the multi hypothesis prediction stage is a challenging task. To resolve this problem, this paper constructs a novel compressed sensing-based high-quality adaptive video reconstruction optimization method. It mainly includes the optimization of prediction blocks (OPBS), the selection of search windows and the use of neighborhood information. Specifically, the OPBS consists of two parts: the selection of blocks and the optimization of prediction blocks. We combine the high-quality optimization reconstruction of foreground block with the residual reconstruction of the background block to improve the overall reconstruction effect of the video sequence. In addition, most of the existing methods based on predictive residual reconstruction ignore the impact of search windows and reference frames on performance. Therefore, Block-level search window (BSW) is constructed to cover the position of the optimal hypothesis block as much as possible. To maximize the availability of reference frames, Nearby reference frame information (NRFI) is designed to reconstruct the current block. The proposed method effectively suppresses the influence of the fluctuation of the prediction block on reconstruction and improves the reconstruction performance. Experimental results show that the proposed compressed sensing-based high-quality adaptive video reconstruction optimization method significantly improves the reconstruction performance in both objective and supervisor quality.
engineering, multidisciplinary,mathematics, interdisciplinary applications