Prior to encoding an input RGB full-color image IRGB, at the server side, performing chroma subsampling on the converted chroma image is a necessary step. After receiving the decompressed subsampled chroma image and luma image at the client side, performing chroma upsampling is also a necessary step for reconstructing the RGB full-color image. In this paper, we consider seven commonly used chroma subsampling methods, denoted by Cs, and four widely used chroma upsampling methods, denoted by Cu. For each combination cs-cu in CsxCu, we first utilize the moment balance law to analyze the coordinate displacement (CD) bias problem occurring in cs. Next, for the combination cs-cu, we analyze the CD bias problem occurring in the transition from the server side to the client side. Then, we explain why the CD bias problem degrades the quality of the reconstructed RGB full-color images in the current coding system. To remedy this CD bias problem, a CD compensationbased (CDC-based) quality enhancement method is proposed to improve the quality of the reconstructed images. To the best of our knowledge, this is the first work in this research direction. Based on the IMAX, Kodak, SCI (screen content images), and Video datasets, the comprehensive experimental results have demonstrated that on the newly released versatile video coding (VVC) platform VTM-12.0, the proposed CDC-based quality enhancement method in our augmented coding system can achieve substantial quality improvement for 17 combinations in CsxCu.

Traditionally, prior to compressing an RGB full-color image, for each converted 2x2 CbCr block BCbCr, chroma subsampling only downsamples BCbCr, but without changing the luma block BY at all. In the current research, a special linear interpolation-based, namely the COPY-based, chroma subsampling-first luma modification (CSFLM) study has attempted to change the luma block for enhancing the quality of the reconstructed RGB full-color image. In this paper, a fast and effective nonlinear interpolation, namely the bicubic convolution interpolation (BCI), based iterative luma modification method for CSFLM is proposed. In our iterative method, a BCI-based distortion function and its convex property proof are first provided. Next, based on the proposed convex distortion function, a pseudoinverse technique is applied to obtain the initial luma modification solution, and then an iterative method is proposed to improve the initial luma modification solution. Based on five testing image datasets, namely the IMAX, Kodak, SCI (screen content images), CI (classical images), and Video datasets, the thorough experimental results have demonstrated that on the newly released Versatile Video Coding (VVC) platform VTM-12.0, our iterative luma modification method achieves substantial quality, execution-time, and quality-bitrate tradeoff improvements when compared with the existing state-of-the-art methods.

Setting a fixed pruning rate and/or specified threshold for pruning filters in convolutional layers has been widely used to reduce the number of parameters required in the convolutional neural networks (CNN) model. However, it fails to fully prune redundant filters for different layers whose redundant filters vary with layers. To overcome this disadvantage, we propose a new backward filter pruning algorithm using a sorted bipartite graph- and binary search-based (SBGBS-based) clustering and decreasing pruning rate (DPR) approach. We first represent each filter of the last layer by a bipartite graph 𝐾1–𝑛, with one root mean set and one 𝑛-weight set, where 𝑛 denotes the number of weights in the filter. Next, according to the accuracy loss tolerance, an SBGBS-based clustering method is used to partition all filters into clusters as maximal as possible. Then, for each cluster, we retain the filter corresponding to the bipartite graph with the median root mean among 𝑛 root means in the cluster, but we discard the other filters in the same cluster. Following the DPR approach, we repeat the above SBGBS-based filtering pruning approach to the backward layer until all layers are processed. Based on the CIFAR-10 and MNIST datasets, the proposed filter pruning algorithm has been deployed into VGG-16, AlexNet, LeNet, and ResNet. With similar accuracy, the thorough experimental results have demonstrated the substantial parameters and floating-point operations reduction merits of our filter pruning algorithm relative to the existing filter pruning methods.