BVI-DVC

Deep learning methods are increasingly being applied in the optimisation of video compression algorithms and can achieve significantly enhanced coding gains, compared to conventional approaches. Such approaches often employ Convolutional Neural Networks (CNNs) which are trained on databases with relatively limited content coverage. BVI-DVC is a new extensive and representative video database for training CNN-based coding tools, which contains 800 sequences at various spatial resolutions from 270p to 2160p. Experimental results show that the database produces significant improvements in terms of coding gains over three existing (commonly used) image/video training databases.

Please note that due to copyright restrictions, this dataset cannot be released. An open dataset, BVI-DVC Part 1, is available at https://data.bris.ac.uk/data/dataset/3h0hduxrq4awq2ffvhabjzbzi1. It comprises a portion of the original dataset (totalling 772 video files).

深度学习方法正越来越多地应用于视频压缩算法的优化，相较于传统方法，可实现显著更高的编码增益。此类方法通常采用卷积神经网络（Convolutional Neural Networks, CNNs），但这类网络往往在内容覆盖范围相对有限的数据集上进行训练。BVI-DVC是一款全新的大规模且具有代表性的视频数据集，用于训练基于卷积神经网络的编码工具，该数据集包含800段不同空间分辨率的视频序列，分辨率范围从270p至2160p。实验结果表明，相较于三款现有通用图像/视频训练数据集，该数据集可在编码增益方面实现显著提升。