摘要
为缩短视频编码和解码的时间,提升编码效率,引入深度学习算法,开展对视频编码技术的设计研究。首先,通过视频帧内预测和帧间预测,去除视频序列时间域冗余信息;其次,利用深度学习算法,对视频进行环路滤波处理;最后,生成高性能参考帧,并利用参考帧完成编码。通过对比实验证明,新的视频编码技术可有效缩短视频编码和解码的时间,从而提升编码效率,减轻视频资源存储和传输的负担。
In order to shorten the time of video encoding and decoding,and promote the improvement of encoding efficiency,deep learning algorithms are introduced to carry out design research on video encoding technology.Firstly,through video intra-frame prediction and inter-frame prediction,the redundant information in the time domain of the video sequence is removed.Secondly,the deep learning algorithm is used to perform loop filtering processing on the video.Finally,high-performance reference frames are generated,and the reference frames are used to complete the encoding.Through comparative experiments,it is proved that the new video coding technology can effectively shorten the time of video coding and decoding,and achieve the purpose of improving coding efficiency,thereby reducing the burden of video resource storage and transmission.
作者
罗雪
LUO Xue(School of Big Data and Artificial Intelligence,Zhengzhou University of Science and Technology,Zhengzhou Henan 450000,China)
出处
《信息与电脑》
2022年第23期194-196,共3页
Information & Computer
关键词
深度学习
视频编码
资源存储
deep learning
video encoding
resource storage
