Video perception model coefficient calculation method, video perception evaluation method and device

The invention discloses a video perception model coefficient calculation method and device and a video perception evaluation method and device, and the method comprises the steps: obtaining a video perception label data set, constructing a cost function, carrying out the gradient descent iterative c...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: ZHANG TAO, HAO LINGLIN, SHI SHUYUAN, JIANG YONG, WANG JIANBIN, ZHAO YU, WANG ZHENKAI
Format: Patent
Sprache:chi ; eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The invention discloses a video perception model coefficient calculation method and device and a video perception evaluation method and device, and the method comprises the steps: obtaining a video perception label data set, constructing a cost function, carrying out the gradient descent iterative calculation of the cost function, and obtaining a target video perception model coefficient, the video sensing model coefficient can be calculated according to a machine learning mode, so that the coefficient calculation accuracy is improved, and the method can be widely applied to the technical field of wireless communication and terminals. 本发明公开了视频感知模型系数计算方法、视频感知评估方法、装置,其中视频感知模型系数计算方法包括获取视频感知标签数据集构造代价函数,通过对代价函数进行梯度下降迭代计算得到目标视频感知模型系数,能够根据机器学习方式对视频感知模型系数进行计算,从而提高系数计算的准确率,可广泛应用于无线通信和终端技术领域。