Posts Tagged ‘ metadata ’

Friday, April 2nd, 2010

by Bob Cutting

ISC West asked ObjectVideo to participate in a panel discussion on the topic of where video analytics will be in the next 10 years.  Past events hosting such a discussion have yielded a somewhat “same old, same old” outcome.  But I thought the discussion at ISC resulted in a much more productive outlook to where video analytics is heading with respect to analytics technology and real world solutions for end users. 

 My key takeaway was a very consistent message from the panel on the age-old question of “where analytics?”—referring to whether we will see more analytics on the edge or on a server.  We all answered in our own way, but as I was listening to all the responses, I realized we were all saying the same thing.  What finally came out is that the question is typically only answered half-way.  People focus more on real-time video processing for event generation.  But there’s a growing solution focus on using those events or underlying metadata in more advanced ways, either for forensic searching or for correlation to other sensor outputs and data streams (POS, ATM, access control…)

So while it’s hard to argue against using edge devices with embedded analytics for video analytic processing and real-time event and data generation, there is a growing solution based on leveraging that data across multiple cameras and locations for advanced business analysis and innovative detection scenarios using multiple data inputs that require back-end, server-based systems to manage the data correlation and detection/search policies.  We’re seeing this exact trend in the market, and these future solutions further justify the value that can be extracted from video analytic technology.

Monday, July 20th, 2009
by Bob Cutting

OK enough is enough.

The next time I attend a conference where that same company gives the same presentation showing how video analytics with color detection could have been used to search for vehicles and prevent events such as the Washington DC sniper killings, I’ll call them out. Using video analytics for color detection is viable in certain applications. But the way this “solution” is presented is just plain careless.

With any video analytic solution, it’s never just about the analytics. It’s about camera coverage…AND the manner in which cameras adjust to light changes…AND the lack of lighting for nighttime applications…AND the blue cast that seems to overlay many objects under certain lighting conditions…AND the sheer fact that customers underestimate the challenge of separating the white/grey/silver/light blue cars that make up 75% of cars on the road! Who has actually sold this and made it work?

This type of non-consultative, haphazard positioning of a solution does a great disservice to our market. Why do vendors continue to compromise our huge collective investment in R&D by making claims based on concept? Would you set up your company for such a customer support nightmare just to win a deal?

Anyone can output good color data as metadata, but environmental influences and consistency of analysis are never fully addressed. Time to educate the market more responsibly and shine the white light of truth on the issue of color detection.