Google launches new Video Intelligence API

At its Google Cloud Next conference in San Francisco, Google announced the launch of a new machine learning API yesterday that can recognize objects in videos and make them searchable.

The new Video Intelligence API is meant for developers to build applications that can automatically detect objects from a video. Until now, image recognition APIs available in the cloud worked only for still images, but with the help of this new API, developers will be able to build applications that let users search and discover information in videos.

“Cloud Video Intelligence API uses powerful deep-learning models, built using frameworks like TensorFlow and applied on large-scale media platforms like YouTube,” Fei-Fei Li, chief scientist for Google Cloud artificial intelligence (AI) and machine learning, wrote in a blog post.

Notably, the API also allows you to tag scene changes in a video. Videos have long been a challenge for machine learning researchers. Google’s two cloud competitors, AWS and Microsoft Azure, do not currently offer this functionality.

In addition, Google also announced that its Cloud Machine Learning Engine, formerly known as Cloud Machine Learning, has now launched out of beta. The same goes for Google’s Cloud Datalab.