摘要
近年来,深度学习技术不仅在人工智能领域取得了巨大成功,也为视频编码领域带来了新的发展机遇。文章从两个方面介绍了深度学习技术在视频编码领域的发展现状,即传统编码框架下深度学习视频编码工具和以深度学习模型为基础的视频编码新框架,并对相关代表性工作进行了详细介绍和性能分析。最后,对深度学习视频编码技术面临的挑战和未来发展方向做了分析和展望。
In recent years,deep learning technologies have not only achieved great success in the field of artificial intelligence,but also brought new development opportunities to the field of video coding.This paper introduces the development status of video coding with deep learning technologies from two aspects,deep learning based video coding tools within traditional coding frameworks and new video coding frameworks based on deep learning models.The related representative works are introduced in detail and their performances are also analyzed.Finally,this paper analyzes and looks forward to the challenges and future development trends for deep learning based video coding technologies.
作者
张新峰
Zhang Xinfeng(University of Chinese Academic of Sciences,Beijing 100049,China)
出处
《信息通信技术》
2020年第2期38-45,共8页
Information and communications Technologies
关键词
视频编码
人工智能
深度神经网络
视频预测
Video Coding
Artificial Intelligence
Deep Neural network
Video Prediction
