Google’s AI can now watch and recognize objects within videos

Google announced a next major step forward for Artificial Intelligence. Fei-Fei Li, the chief scientist of cloud AI, who talks about the current and the next generation application of AI at Google’s Next Cloud Conference.

The company announced the machine learning system that can identify an object in videos. The technology API(Application Programming Interface) that is implemented so-called Cloud Video Intelligence.

Now, it is only available as a beta to a Cloud developer group.

On Twitter, Karissa Bell tweeted about the new video Intelligence API that lets you search for objects within video clips. See the video clips tweeted by her.

Google just showed off a new Video Intelligence API that lets you search for objects within video clips, here's an bit of the demo pic.twitter.com/wOPwsmnAfe

— Karissa Bell (@karissabe) March 8, 2017

For example, if you search ‘beach’, the technology can find videos which had a scene from that beach. But on youtube, you certainly find video to the related subject.

This kind of services primarily users keywords, tags, and description that are given by the user to sort content. Without a human help, cloud video intelligence did a wonderful job.

The Cloud Video Intelligence was developed based on frameworks such as Tensor Flow and then submitted large-scale Database. The incorporated Algorithm can learn to separate the entities(animals, people,.etc) and action(Jogging, swimming,.etc).

By this technology, you will be able to contextualize a searched one if you search ‘tiger’ also shows the related content such as wildlife, zoo, and cat. Cloud Video Intelligence API will appear in the various part of the cloud platform.

However, I will not be surprised if Google harnesses this technology on YouTube, but at least in the initial phase, the API will only be available to organizations using the Google Cloud Storage service (including companies like Evernote and Spotify).