Yup, you heard that right. At its Cloud Next developer conference that is currently taking place in SF, CA, Google is revealing all sorts of good stuff. The company has announced the launch of a brand new machine learning tech that can enable automatic recognition of objects from inside videos.
Announcing the service, Fei-Fei Li, Chief Scientist, Google Cloud AI and Machine Learning, said:
Cloud Video Intelligence API (now in Private Beta) uses powerful deep-learning models, built using frameworks like TensorFlow and applied on large-scale media platforms like YouTube. The API is the first of its kind, enabling developers to easily search and discover video content by providing information about entities (nouns such as “dog,” “flower” or “human” or verbs such as “run,” “swim” or “fly”) inside video content.
What’s more, the API can also make these objects searchable. This is due to the image extraction tech Google has built right within the the new Video Intelligence API. Developers can deploy the API to automatically get information extracted from videos.
The tech has already been around, however, its the first time that it is being deployed for this purpose. In case you are wondering where you encountered it before: Well, doesn’t image search allow you to do something very similar? Only that particular process is for pictures alone.
So with this, developers will be able to create applications that allow the user to search using generic terms and then come up with results — in form of videos that have whatever the user searched for. So, for instance if the user searched for dog or fruit, the search result will consist of videos that have these elements in it.
The API also lets you actually tag the scene changes in the video. Meanwhile, these videos must be stored in Google’s cloud storage service.
See, images and videos are not really all that different right? Both are made up of pixels. However, videos have many many images displayed in rapid succession and that tends to complicate things quite a bit. Now though, Google has found its way around that particular corner.
Google also announced that the cloud Machine Learning Engine, which is a tool for building custom machine learning models using Google’s TensorFlow framework, is now generally available.
You can take a look at the workings of the system here. Developer can sign up for the private beta here.