机器学习在HEVC编码实现中的应用
首发时间:2015-11-30
摘要:机器学习能帮助挖掘出数据中隐藏的相关性等信息,因而开始应用在视频编码领域。本文对机器学习在HEVC编码实现中的具体应用进行了综述。首先介绍了包括朴素贝叶斯,SVM以及决策树算法在内的部分经典机器学习算法,并对每种机器学习算法的适用范围进行了分析。然后归纳了在编码实现中的分类,聚类,参数训练等类型的问题可以使用机器学习方法,从而提高编码实现速度或压缩效率。
For information in English, please click here
Machine learning in HEVC encoding implementation
Abstract:Machine learning can help dig out the hidden information, such as correlation, therefore begins to be applied in the field of video coding. This paper overviews specific applications of machine learning used in HEVC encoding implementation. Firstly, classic algorithms including naive Bayes, SVM, decision tree as well as applicable scope of each algorithm are analysed. Then three question types (classification, clustering and parameter training) are generalized in encoding implementation, which can be solved by machine learning. The result shows that machine -learning based method can decrease coding time or improve compression efficiency in HEVC encoding implementation greately.
Keywords: multi-media machine learning fast decision
论文图表:
引用
No.4665020111657514****
同行评议
共计0人参与
勘误表
机器学习在HEVC编码实现中的应用
评论
全部评论0/1000