Over the past two days, I've spent quite a bit of time with Stevie Clifton, Co-founder & CTO of Animoto. Besides being one of the coolest things I've seen in years, Animoto is giving us a glimpse of automated imagery PED (Processing, Exploitation, Dissemination). First an introduction.
Animoto Productions, a self described "bunch of techies and film/TV producers who decided to lock themselves in a room together and nerd out" have released a web application that automatically generates professionally produced videos. The site uses their patent-pending technology and high-end motion design to fully customized orchestration of user-selected images and music. By using Cinematic Artificial Intelligence technology, "it analyzes and combines user-selected images and music with the same sophisticated post-production skills & techniques that are used in television and film." Their AWS spike story is now famous in the cloud computing computing.
Now let's fast-forward 5, no, 2 years. A UAV is now flying over the US southern border streaming live video to an intelligence center. Using frame-grabbing technology, it forwards a series of still images to the automated intelligence exploitation center. The face recognition program matches one of the images to a known smuggler, which kicks of an automatic query of NCIC and NLETS. Timestamp information is also used to create an audio track from the many high fidelity microphones in the area. The audio, still frames and automatic query data is then sent to the Animoto engine, which uses the available meta-data to produce a intelligence video and transmits it, in near-real-time, to the nearest CBP unit for appropriate interdiction.
By the way, Animoto uses Amazon Web Services with Rightscale to provide it's service.