BVI-DVC Part 1

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 772 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.

深度学习方法正日益广泛地应用于视频压缩算法的优化工作，相较于传统方案，可获得显著提升的编码增益。此类方法通常采用卷积神经网络（Convolutional Neural Networks，CNNs），而这类网络往往在内容覆盖范围相对有限的数据库上完成训练。BVI-DVC是一款全新构建、覆盖广泛且具有代表性的视频数据库，专为训练基于卷积神经网络的编码工具而设计，该库包含772段具备从270p到2160p多种空间分辨率的视频序列。实验结果表明，相较于三款现有（常用）的图像/视频训练数据库，该数据库可在编码增益方面实现显著提升。