Our latest webinar on TMCnet was called Voice and Video over IP Communications: Assessing and Improving User Experience.
As we already know that video calling is happening and we know how it fits into the UC world, we decided to discuss on this webinar the different issues related to the quality of video calling. The focus was around video, although we did deal with audio as well.
For video, there are two main approaches for assessing and improving the quality of experience:
- Assessing quality using reference, which means sending a specific video content through the system and then checking the received video (the result) against the reference.
- Assessing the quality with no reference at all, which means evaluating the quality of the video received, without any reference video to rely on.
In the webinar we tried to explain the differences between the two, along with the techniques we use and the algorithms we employ in our new video quality assessment tools – VQ SDK and VQ Monitor. For me, the interesting part in the webinar was the explanation given by my colleague, Tamar Barzuza, a Senior Algorithm Researcher, on the techniques that can be employed in order to evaluate video quality without reference.
I believe that video quality is a key issue for any video product developer. It is very important for most of the people who attended our webinar – over 50% already have some kind of a video implementation in their product.
Another interesting poll question we asked during the webinar was about the factors that affect the quality of real-time video:
I’d say that our audience pointed at the exact same issues we feel are the most important – packet loss and bandwidth.
You can view this media quality webinar online if you want.
I have also embedded here the slides for your enjoyment:
