Research on quality assessment of network video based frame freezing duration

doi:10.1145/3349341.3349492

	Research on quality assessment of network video based frame freezing duration
	Zhao, Hong; Kong, Dongyi; Cao, Chang; Ma, Donglin
	2019-07-12
会议日期	July 12, 2019 - July 13, 2019
会议地点	Wuhan, China
关键词	Artificial intelligence Mean square error Video recording Video streaming Linear correlation coefficient No references Payload information Root mean square errors Temporal information Video frame Video quality Video quality monitoring
DOI	10.1145/3349341.3349492
页码	691-697
英文摘要	Delivery service of high quality video streaming is a big challenge for providers of network video due to there is huge video data need to be delivered in time and the network condition is complex and dynamic, which have a severe impact on fluency of playing video. The frame freezes as one of the main factors affects the fluency of playing video severely. A novel model of no-reference network video quality assessment is proposed to improve the process of video quality monitoring. The assessment model is constructed according to frame freeze duration, concentrative degree of frame freeze and complexity of video frame content. The frame freeze duration and concentrative degree of frame freeze are acquired by frame rate; the number and location of frame freeze events are obtained from analyzing the frame freeze events occurred in playing video. The complexity of video frame content is calculated by coding bit-rate, the type of macro-block and the motion vector, which obtained from payload information of video streaming. The tests are performed on a set of video sequences with different spatial and temporal information. Compared with G.1070 model, the Pearson Linear Correlation Coefficient of the proposed model is increased by average of 0.041, and the Root Mean Square Error reduced by average of 0.049. Test results indicate that the proposed model has a better performance in assessing frame freeze in network video. © 2019 Association for Computing Machinery.
会议录	ACM International Conference Proceeding Series
会议录出版者	Association for Computing Machinery
语种	英语
内容类型	会议论文
源URL	[http://ir.lut.edu.cn/handle/2XXMBERH/117967]
专题	计算机与通信学院党委学生工作部（学生处、易班、武装部）
作者单位	School of Computer and Communication, Lanzhou University of Technology, Lanzhou, China
推荐引用方式 GB/T 7714	Zhao, Hong,Kong, Dongyi,Cao, Chang,et al. Research on quality assessment of network video based frame freezing duration[C]. 见:. Wuhan, China. July 12, 2019 - July 13, 2019.