At its Cloud Next conference in San Francisco, Google today announced the launch of a new Auto-Learning API for automatic object recognition in videos and their search. Google Learning API recognizes objects in videos
The new Video Intelligence API will allow developers to create applications that can automatically extract features from a video. So far, most similar image recognition APIs available in the cloud have only focused on doing this for still images. But with the help of this new API, developers will be able to create applications that allow users to search and discover information in videos. That means you can look for “dog” or “flower”, for example.
In addition to extracting metadata, the API allows you to mark the scene changes in a video.
According to Fei-Fei Li, his main AI scientist and Machine Learning in Google Cloud. In today’s speech, the world of pixels goes beyond images. Videos have long been a challenge for machine learning researchers. This new service, however, now makes extracting information from these videos as easy as doing the same for images.
In addition, the Cloud Machine Learning Engine, is the company’s tool for building custom machine learning models using its Tensor Flow framework.
As Li pointed out in today’s speech, the company wants to democratize the auto-learning technologies it has developed internally. The Vision API is another example of this.
The goal of creating increasingly intelligent machines is met little by little. With the help of powerful programs a car can travel without the management of a human. Find and reserve for you a table in a restaurant or accompany the chords of a pianist. But one of the most surprising functions is to recognize the objects of an image. Although Google has presented an app to recognize the objects that appear in the videos.