There was fascinating article in the NY Times about video/content recognition software - a utility typically used for video surveillance systems.
This article showcased an interesting twist on its use...Cracking Down on, and preventing Online Video Piracy (DRM) !
Essentially, this California company demonstrated its software running while viewing YouTube. One clip of an illegal movie upload, even while distorted, in black-and-white, and overdubbed from a video pirate, was immediately and correctly identified - by its exact scene.
Hearkening back to the Sony Root-Kit debacle, I shudder to think what a big media company wouldn't give to be able to distribute a root-level application to users that "watches and listens" to content to make sure it's been validated. If Vista checks up on your drivers a gazillion times a minute, it's certainly not too much to imagine it checking up on your media too.
Showing posts with label index. Show all posts
Showing posts with label index. Show all posts
Wednesday, February 21, 2007
Sunday, October 15, 2006
Late Night Thoughts About an Index Really Worth More Than a Thousand Words
What if a photograph could be, in of itself, a search query?
A photo could be uploaded and links with descriptions fed back based on what the algorithm detected the photo to be. The perfect match might tell you who an unknown person is, or when the photo was taken. Perhaps it could detect and reference location based on skyline patterns or landmark geometry. Perhaps similar photos could be be fed back, giving say, more photos of the same person or place.
What if our Voice communications, like IP telephone conversations, were recorded and indexed and could be searched like emails are with Gmail? Not in a "big brother" sense, but say for mutually accepted and practical reasons. I think of business conversations, teleconferences, and presenations.
It seems so rudimentary, so simple, but its very powerful and some might say frightening: When everything is completely digital, it can all be, at least theoretically, indexed and searchable. It's certainly not out of the realm of possibility that algorithims could be developed to do it. That kind of index would really be worth more than a thousand words.
A photo could be uploaded and links with descriptions fed back based on what the algorithm detected the photo to be. The perfect match might tell you who an unknown person is, or when the photo was taken. Perhaps it could detect and reference location based on skyline patterns or landmark geometry. Perhaps similar photos could be be fed back, giving say, more photos of the same person or place.
What if our Voice communications, like IP telephone conversations, were recorded and indexed and could be searched like emails are with Gmail? Not in a "big brother" sense, but say for mutually accepted and practical reasons. I think of business conversations, teleconferences, and presenations.
It seems so rudimentary, so simple, but its very powerful and some might say frightening: When everything is completely digital, it can all be, at least theoretically, indexed and searchable. It's certainly not out of the realm of possibility that algorithims could be developed to do it. That kind of index would really be worth more than a thousand words.
Subscribe to:
Posts (Atom)