Tech Features

Mining the AI potential: Jon Morgan of Object Matrix

Jon Morgan is CEO of Object Matrix.

AI search will monetise archives in new ways and AI data management will move data and data processing algorithms to where they need to be, writes Jon Morgan of Object Matrix.

Once upon a time, there was an industry that called programmatic code ‘software’, not artificial intelligence. In media storage you’ve probably heard AI used in the context of image searching your media archive to find a matching actor’s face. Because AI has become an almost ubiquitous phrase, and therefore somewhat grating, it would be easy to dismiss it from our thoughts – but is there anything behind the term that we should be taking notice of?

Search: The final frontier

Google processes a lot of data. Some estimates say Google handles 1.2 trillion searches per annum. I don’t know, but I’d imagine 99.999% of those searches are based around language. However, in a world where an estimated 80% of data is unstructured and the vast majority of that unstructured data is video data, it’s clear that search has a long way to go.

The next stage is to be able to search video and, for reasons we will briefly touch on, the techniques used to analyse and allow for the meaningful search of video are often called AI.

Why AI? A few years ago, we would have called it image searching, video searching or software algorithms, but since the success of Apple Siri, Amazon Echo, etc, the general public psyche and lexicon has begun to consider any such technology to be artificial intelligence – even if it doesn’t have a self-learning feedback loop or some other mechanism of analysis enhancement over time. So, though it irks me, let’s join the bandwagon and call it AI video analysis and search.

What cannot be ignored is that this is exciting. Video archives that previously sat dormant can now be queried for all types of things. What footage of Princess Diana do we have? Have we ever shot a news item on that street before? Who has played number 10 for England?

Those image-searching algorithms are nascent but quickly being realised. On the other hand, semantic video analysis is more basic but will provide more and more answers. For example, this part of the movie is a car chase; look at this CCTV stream, because a street fight has broken out; this movie is a remake of The Magnificent Seven; cars on this road have gone from 100% driven to 30% self-driven in the past five years.

So AI video analysis and search is beginning to unlock the value in media archives, and importantly this allows better reuse of media assets for all types of analysis, documentary making, sports fan look-ups and countless other usages. This will only go from strength to strength.

But is that really what AI in media archiving is all about?

AI: Beyond image searching

The tendency is to think about AI in the context of search, which certainly limits our definition. One of the key challenges with storage today is data gravity – having the video data where and when it needs to be processed – which is a mixture of moving data to where it needs to be and moving processing to where the data is.

Ingenious storage systems are needed to manage this. Take for instance a shoot post-produced by teams in different countries. Assets should be available to those teams as and when they are ready to work on them. That’s where AI intelligence comes in: moving assets around for us.

Furthermore, AI can manage this in a manner that also protects data against any kind of loss, with security in mind. AI storage becomes your 24/7 super-administrator so that you don’t need to spend your days making data transfers or worrying about data security.

The AI in media storage concept should not be purely limited to AI searching; it is also about AI data management.

Can video search change industry dynamics for video owners?

All in all, the media industry should be hugely excited about the new video search techniques coming on to the market, for the simple reason that they allow holders of media archives to better monetise and re-use content.

But could this be just the beginning? Will AI production mixed with video search be able to cut new content? Will video search allow a whole new analytics industry to develop – new Googles, new Facebooks? Will the owners of video archives gain more of their revenue from them than they do from new content?

The future is uncertain, but the foundations of that future are here and now.

The term AI will quickly become redundant (again), since it is being used too ubiquitously! However, AI search will include image search, object search, semantic analysis and speech to text. It will revolutionise how and why we hold video archives. It will monetise archives in new ways – keeping archives offline will no longer be viable. It will move from niche to mainstream over the coming years.

AI data management will move data and data processing algorithms to where they need to be and will automate the protection and security of assets. The future potential for media archiving is more exciting now than it has ever been.