Posts

MSU Subjective Codecs Comparison 2017 Released

MSU Video group of Graphics & Media Lab has published Video Codecs Subjective Comparison Report 2017 devoted to comparison of modern video encoders according to subjective visual quality of encoded streams they produce. The report compares 7 software video codecs:

Read more Dec 1, 2017
Replication of Netflix Subjective Study

Undoubtedly, subjective study conducted in well-controlled laboratory environment with huge panel of observers is the gold standard for comparison of image and video processing algorithms. Unfortunately, crowdsourced studies do not allow researchers to control content viewing conditions, since study participants are viewing content at their homes with various devices. However, it is significantly easier to recruit large amount of participants for crowdsourced study than bring the same number of people to the laboratory. Thus, noise introduced by poorly-controlled viewing conditions of crowdsourced study can be reduced by larger amount of participants.

Read more Sep 6, 2017
Docker for Windows: Watch Bindings

Docker has changed the way how we build and deploy our apps, by letting developers to spawn development or production environment on any (or almost any) machine just by typing docker-compose up. However, initially Docker was not available for Windows users until February 2016 when Docker for Windows was released. Unfortunately, despite support of most features that Docker has and seamless integration with Windows-host's environment, Docker for Windows (DfW) has its limitations. In particular you can't use development server's watch mode of most of modern frameworks (e.g. Jekyll, Angular CLI, etc.), since DfW does not notify container about any file changes you make from Windows. In this post we will describe workaround for this problem.

Read more Apr 24, 2017
Subjective Comparison of Video Matting Methods

In this post we briefly describe how we used Subjectify.us platform to carry out subjective comparison of video and image matting methods. In the last section we show study results and discuss them.

Read more Apr 15, 2017