The past two years have seen dramatic advances in hardware, firmware, software, mobile communications and new tools for mining unstructured data that can add value to security and business. Photo: ©istockphoto.com
Different elements that were required to make video content analysis (VCA) stable, reliable and viable have finally caught up with us. Having embraced this exciting concept back in 2005, I was overly optimistic and seriously unarmed with the tools that ultimately would make this technology useful for both security and business purposes.
The past two years have seen dramatic advances in hardware, firmware, software, mobile communications and new tools for mining unstructured data into useable nuggets that can add value to security and business. Most importantly has been a fundamental shift in attitude from “my proprietary product” to “collaborate to expand the market.” The NIH (not invented here) mindset is changing for the better. Perhaps it is the influence of other technology segments that view the world differently or perhaps just a sign of our changing times.
Does 2013 promise to be a break out year in the development and deployment of video content analysis and facial analytics? It’s a pretty deep question to ponder all by myself so I threw out a lifeline to gain some different perspectives. I called two knowledgeable visionaries of video/facial analytics that service two very distinct market segments for some help. What follows is the first of consecutive columns devoted to pinpointing the present state of video analytics.
DoD Drives Analytics Development
Paul Schuepp is the CEO and visionary of New Hampshire-based Animetrics, and shares my Swiss and engineering background. OK, he is a much better engineer than I ever was. The firm serves the government/law enforcement/military segments, and agencies with lots of initials that are very interested in “persons of interest” so we all can live in a safer world, including our frontline soldiers. The DoD was a demanding customer and driving force for field deployment of workable video analytics that did not enjoy “ideal environments” but would save lives. Animetrics responded to the challenge.
The DoD needed to more accurately analyze facial data and recognition in the less-than-ideal conditions outside of conventional facial capture parameters. The ability for military or law enforcement capture of facial details, then to be able to compare them to a wide variety of databases, represents a new era in this technology category. This technology can analyze captured or live images with a +/-45° frontal facial angle.
That is a big step forward. Add the fact that this technology can work with field-generated, captured static facial images from smartphones and voila, reasonably real-time and actionable data. Mobile communication and video devices have now increased the range and flexibility of deployment, making the underlying software advances even more valuable as a security tool.
Schuepp shared plans for making some of the company’s proven and patented technology available in SDK and API tools in the near future. Untethered high quality mobile image capture (smartphone) and fast inexpensive transmission of data to serious, cloud-based processing power is a reason that video analytics will start delivering on its real potential. Multiply the different law enforcement or military databases those pictures or video may reside in our expanding government and you get BGD — Big Government Data.
Managing Data Overload
The acceptance and adoption of cloud-based processing power that is kept state-of-the-art by the provider, but sold on an as-needed basis, is going to make all the difference. The ability to throttle up technology support tools on demand and substantiated by immediate need makes good business sense for a customer.
If you mix in the right technology applications enabled by open SDK and APIs with the continued advancement price/performance ratios of edge appliances (cameras or smartphones), you have the recipe for security and business intelligence that is manageable, actionable and affordable. In the military and business we call this a force multiplier.
Al Shipp is the CEO and visionary force behind San Francisco’s 3VR, which serves the commercial/retail/banking market segments — arguably a tougher crowd to penetrate, sell and please for any integrator that lives in that space. Shipp, who also contributed to this article, has brought old-school engineering disciplines (IBM) and new school technology thinking (Apple) experience to 3VR for the past five years.
The firm pioneered the commercialization of the video content/facial analytics technology movement into video analytics with clients that were early adopters. Its solutions deliver “real-time info on real-time video” that allows those in security and marketing across the enterprise to quickly recognize the significance of a predefined event, object or person, and then act immediately.