Don’t let bad apples spoil your video goods – SSIMPLUS

How much are bad video sources costing your streaming service? Sure, you can re-process them if you catch them before delivering to your customers. But what if you have 10 new assets to debut on your VOD service today? Or if you have a real-time sports delivery? How much will a bad source cost you if a compromised video reaches 1 million, 10 million or even 50 million customers? Are you eager to take that risk and create negative word of mouth, reputational damage and even churn?

Data based on more than 10 million hours of video shows that ~50% of sources fall below the SSIMPLUS Viewer Score of 90 which is a benchmark for premium quality. By the time these videos reach subscribers, the Viewer Experience will have deteriorated below 75 when a Viewer Score of 80 is needed to keep subscribers happy and engaged.

So how do you make sure that there are no rotten apples in your video basket?

SSIMWAVE Source Validation solution provides the most accurate perceptual quality measurement for no-reference situations and ensures that your source content is worth your processing. It helps to:

• Keep content partners accountable for what they deliver;
• Catch hard to detect pixel-based impairments like banding, macroblocking, compression artifacts and upsampling;
• Minimize waste of valuable resources on processing videos that will only get worse with each step;
• Protect and grow subscriptions by maintaining quality thresholds and decrease churn by stopping bad video in its tracks;
• Save valuable time by helping workflows work seamlessly 24/7.

SSIMPLUS is the only perceptual video quality metric to work in situations where the reference source is not available to measure against.

VMAF, SSIM, PSNR simply don’t work for two reasons:

1) They are full-reference metrics, and

2) VMAF and SSIM are not real-time, which is crucial for LIVE delivery.

At the same time, approaches that can catch video quality issues like letterboxes or interlacing are unable to quantify the impact of these issues on the Viewer Experience, so you get the data with no clear indication of what to do with it.

SSIMWAVE’s no-reference Source Validation solution is based on more than 30 years of research into the Human Visual System and Emmy Award-winning algorithm. SSIMPLUS Viewer Score uses a linear 0-100 scale that serves as the main KPI to keep your content partners accountable. It can even facilitate optimization decisions based on the SSIMPLUS Viewer Score—knowing the quality for each source can help you identify sources with higher quality than needed. Why send an animation at a 90 score when an 80 will reduce costs and still look good?

Our experience of processing more than 10 million hours of video each month shows that ~50% of sources fall below the SSIMPLUS Viewer Score of 90 which is a benchmark for premium quality. That score is even lower for sports and Prime time content. Our data also shows that by the time a video reaches the subscriber, the Viewer Experience will have deteriorated below 75. We know from our extensive Viewer Experience research that to have happy subscribers, a minimum of an 80 Viewer Score at their device is needed.

SSIMPLUS is simple to use, deployed within a SAAS model, and features virtual probes that integrate in any VOD/LIVE workflow. It provides Viewer Intelligence® with clear insights:

• Which sources/assets failed;
• How frequently per time zone/day/month/year;
• How often do certain providers fail;
• Ranking of content providers by quality to help in SLA and price negotiations; and
• Source quality deltas vs. the targeted Viewer Experience score thresholds.  BCS Bureau