User network model and QoE

”

8 month 5 Japan -6 Japan ,LiveVideoStackCon 2022 Audio and Video Technology Conference Shanghai Station , Open the door to the future with you .

User network model and QoE

In audio and video applications , a Get a lot of data reported by users , Including but not limited to audio and video quality data 、 User behavior data, etc , What conclusions can these data provide us ？ Whether these data can be used to establish a model for rapid experiment ？ Can we use this model fast iteration strategy to improve online audio and video quality ？

Lecturers and topics

Facing the large-scale distributed transmission system covering the world , In the face of complex live broadcast 、 Call scene , Alibaba cloud is based on flexible tuning methods 、 efficient AB test Basic capabilities and observation and analysis methods close to business scenarios , Summarize a set of iterative tuning methods , This sharing will introduce the global real-time transmission network in detail GRTN QoS Optimization Practice .

Audio and video quality inference is collected through the whole link index 、 Network simulation 、 The ability of three parts of quality data analysis to achieve , From all dimensions 、 Measure the quality of audio and video communication in each stage and scene 、 Compare the quality change trend of each version 、 Provide solutions to the quality improvement of audio and video . By analyzing the results, we can help developers find problems in time 、 Positioning reason 、 And solve problems efficiently , To improve the operational efficiency of customers and user experience , It can also guide the operation deployment , Guide routing strategy and collection 、 Play 、 codec 、 Transmission and other policy adjustments .

This sharing will be divided into three parts , The first part introduces the whole link index collection of huanju group and how to ensure the accuracy of index data ; The second part introduces the automatic simulation tool , How to improve test efficiency 、 Version quality comparison 、 Auxiliary algorithm optimization and improvement, etc ; The third part introduces the quality data analysis , How to realize the analysis and comparative analysis of single index and multi index .

Human perception oriented quality evaluation plays a very important role in many video image processing algorithms and systems . In recent years, many quality evaluation methods have been put forward in academic circles , High performance has been achieved on existing data sets , However, their performance in practical application is still not satisfactory to users , So that it cannot be widely used .

In this report, we will review the research paradigm of traditional visual quality evaluation , This paper emphatically points out a problem that has been neglected for a long time in the field of visual quality evaluation ： On an existing dataset , The statistical results of coarse-grained quality evaluation mask the evaluation of fine-grained quality differences . Accordingly , From the perspective of fine-grained quality evaluation , This paper briefly introduces a small number of fine-grained quality evaluation studies at present , It also looks forward to the opportunities and challenges faced by fine-grained quality evaluation in the future .

 

---

In the scan QR code Or click on Read the original

Learn more about the conference

If you like our content, just order “ Looking at ” Well ！

This article is from WeChat official account. - LiveVideoStack（livevideostack）.
If there is any infringement , Please contact the [email protected] Delete .
Participation of this paper “OSC Source creation plan ”, You are welcome to join us , share .