There’s no doubt that searching video footage can be an incredibly tedious and time consuming task — unless someone else has manually tagged a video with keywords, there’s a good chance you’ll have to wade through minutes (or hours) of footage to find what you’re looking for. In the last few years we’ve started to see the emergence of some new technologies to address this issue, like Digital Smiths, but most of them are focused indexing premium content. Vitamin D is a new startup making its debut today at DEMO Fall that’s looking to bring content identification to user generated videos.
Vitamin D’s first product allows users to search through surveillance camera videos without having to watch though hours of footage where nothing is happening. But this isn’t just based on motion detection — you can fine tune your searches using a visual interface for building a custom ruleset. Say, for example, I wanted to monitor a specific door in a room for any humans that had walked through it. Using Vitamin D, you can drag and drop to highlight a certain doorframe and then ask the software to immediately identify any matches it has in its archive. Using this technology you could condense many hours of footage into only a few seconds, which you can browse through a in playlist-like format. The company says it will have a public beta available this fall. Vitamin D also looks to apply its technology beyond its security application, though it sounds like further products are a ways off.
Also see HighlightCam, which can automatically condense lengthy video clips to their most important moments. The startup was a part of Summer 2009’s Y Combinator class (you can find our full post on HighlightCam here).
