After patenting the ability to automatically detect faces in videos, Google has now been granted a patent for automatic detection of objects in videos. In other words, Google now has the rights to the technology that allows for automatically detecting and tagging objects in videos (whether that be a living people, inanimate objects, or animals).
Similar to how it plans on implementing automatic face detection, Google wants to create a database of “feature vectors” which can be drawn upon to detect objects in videos. This database would house characteristics of objects such as like color, texture, shape, movement, etc.; videos would be analyzed, objects pinpointed based off the characteristics stored in Google’s database, and as tagged as such.
To help improve detection and tagging accuracy, users can manually tag objects in videos to let Google know what is what. Google will use this manual tagging as a fall-back in the situation when Google’s automatic tagging fails, and Google may collect the manual tagging data to improve the data in the underlying database.
Of course the obvious use of this new patent would be on Youtube where millions of videos are viewed a day. More interesting, however, is the possibility that this technology could eventually find its way into Project Glass, a next-generation augmented reality project Google has been working on for some time now. That would be damned cool, if I do say so myself.