Plenty of unfamiliar faces were jostling among the cable industry old guard at Anga Com this week, few hotter right now than SSIMWave – the Canadian perceptual quality start-up which has just unveiled its new VoD Monitor Inspector tool for video processing and delivery optimization. The launch doesn’t sound like much initially, until you remember SSIMWave is claiming a clean sweep of trials at North American tier 1 operators with public deployments potentially just around the corner.
When Faultline Online Reporter got stuck into conversation with SSIMWave just a couple of months ago, the emphasis was largely on live streaming and how the main benefit of the company’s SSIMPlus technology over the competing metric VMAF (Video Multimethod Assessment Fusion), developed by Netflix is that VMAF cannot handle live. So, the launch of a new VoD-centric add-on product to its flagship SSIMPlus technology, at a show where the delivery of live content is such a major focus point, will surely have raised some eyebrows in Cologne – particularly when flaunting the industry’s only algorithm to achieve machine-to-human correlation accuracy exceeding 90%.
Aimed at video engineers and architects, VoD Monitor Inspector is designed to sculpt more effective encoding and processing workflows for on-demand content – aiding customers in deciding on configuration and purchasing. The technology evaluates HD video in real time, SD video at 4x speeds, and 4K at unspecified “unprecedented” rates. But while the SSIMPlus technology has long been processing on-demand content, the point of VoD Monitor Inspector seems to be about making the technology more readily available for video professionals. And with VoD Monitor Inspector being the first in a suite of products expanding on its flagship technology, it seems the technology is poised to become even more accessible.
More interesting still is how SSIMWave did not view Netflix’s VMAF as a competitor two months ago, but here we have a new product which we see as directly targeting the open source VMAF user base. In addition, SSIMWave has been gaining ground more recently (in terms of efficiency score tests) by focusing on comparing individual components like encoders, transcoders and packagers.
VMAF, meanwhile, looks at individual quality metrics like visual information, detail loss and motion (temporal difference between adjacent frames), which only provides partial measures of perceptual quality, although can be amalgamated for better accuracy. Netflix has applied conventional machine learning to assign weights to different metrics and tune the model until it matches human perception as closely as possible.
The latest version of VMAF featured at NAB 2019 is a product called ClearView from Video Clarity, which has found it very consistent across content types and is adept at assessing artefacts associated with streaming.
That said, why would a company pay for VoD Monitor Inspector when VMAF is free and open source? “SSIMPlus is a paid product but it’s winning, so we must be doing something right,” was the simple and staunch defense of SSIMWave’s VP Marketing Saj Jamal, speaking to Faultline Online Reporter recently.
Naturally, VoD Monitor Inspector is powered by SSIMWave’s core Structural Similarity technology, based on algorithms we learned recently have been in development for around 30 years, essentially encompassing the accuracy of hundreds of thousands of human viewers – data which is then used to apply scores to video content. The new cloud-based VoD Monitor Inspector tool combines frame-by-frame and pixel-by-pixel analysis of content using the core SSIMPlus algorithm. The scoring system is as follows: Excellent (81-100), Good (61-80), Fair (41-60), Poor (21-40), and Bad (1-20) – whereby a score of 80+ is generally considered the equivalent of an HD TV broadcast.
SSIMWave is designed to delve further into video analytics compared to run of the mill QoE metrics like bitrate, frame rate and network conditions. It can score factors based on specific device brands and models, be it size of the screen, distance from the screen (based on sound), brightness, number of pixels, and it can even acknowledge that viewing video on a smartphone likely means the handset is tilted at approximately a 30-degree angle – factoring this into its decision making. Device manufacturers are now asking SSIMWave if this QoE viewing data is important to carriers and these carriers are increasingly seeing the importance of this data and in turn are more willing to pay for the privilege.
Right now, Telefonica is the only tier 1 operator customer SSIMWave can discuss openly but the vendor recently suggested something significant is on the horizon.