Tree-Structured Data Clustering-Driven Neural Network for Intra Prediction in Video Coding
As a crucial part of video compression, intra prediction utilizes local information of images to eliminate the redundancy in spatial domain. In both the High Efficiency Video Coding (H.265/HEVC) and Versatile Video Coding (H.266/VVC), multiple directional prediction modes are employed to find the texture trend of each small block and then the prediction is made based on reference samples in the selected direction. Recently, the intra prediction schemes based on neural networks have achieved great success. In these methods, the networks are trained and applied to intra prediction to assist the directional prediction modes. In this paper, we propose a novel tree-structured data clustering-driven neural network (dubbed TreeNet) for intra prediction, which builds the networks and clusters the training data in a tree-structured manner. Specifically, in each network split and training process of TreeNet, every parent network on a leaf node is split into two child networks by adding or subtracting Gaussian random noise. Then a data clustering-driven training is applied to train the two derived child networks using the clustered training data of their parent. To test the performance, TreeNet is integrated into VVC and HEVC to combine with or replace the directional prediction modes. In addition, a fast termination strategy is proposed to accelerate the search of TreeNet. The experimental results demonstrate that TreeNet with the fast termination can reach an average of 2.8% Bjontegaard distortion rate (BD-rate) improvement (up to 8.1%) and 4.9% BD-rate improvement (up to 8.2%) over VVC (VTM-4.0) and HEVC (HM-16.9) with all intra configuration, respectively.