Google Recognizes Video Objects Using Video Intelligence API

Fei-Fei Li, Chief Scientist, Google Cloud AI and Machine Learning, demonstrated about the current applications and the next generation of artificial intelligence which will bring major difference in self-driving car and health care as well as search capabilities of the Google Photos.

The new technology “Video Intelligence API” (considering an important presentation of Google) was announced by Fei-Fei Li at Google Cloud’s Conference held in San Francisco yesterday. A special focus was given on this new machine learning API.

Google Cloud Video Intelligence API now lets developers search and discover video content by providing information about entities such as noun like “flower” or “dog” or verb like “run” or “fly” within video content.

For an instance, we make a search for “flower” and it will show you videos that contain flowers along with contextual understanding of those bodies appear.

Here is a good example that clears up the entire process, in which a search has been made for “Tiger” and it showed all precise attempts containing tiger across a video collection.


To use Video Intelligence API, you need to sign up for the account, for that you can simply click on Video Intelligence API to reach the desired platform.


Here you need to provide all required information related to your business. To have more precise insight about it, you can consider given below screenshot:


After filling up all necessary information, click on submit followed by your credit or debit card details as this service is paid for all and thereafter you can start using it.

This new technology is for large media organizations and consumer technology companies who need to develop or find simple ways to manage crowd-sourced content or those who want to built in this feature in their video management software.

This is an addition to existing Google Cloud Machine Learning APIs (Vision, Speech, Natural Language, Translation & Jobs) which allow people to create next generation applications that give ways to parse video content by seeing, hearing and understanding unorganized data. It will definitely expand usage of Machine Learning APIs in every sector from medical image analysis, product recommendations to fraud detection and many more other things.

He also put some light on Cloud Machine Learning Engine that helps you create machine learning models that can handle any sort of data at any scale. It allows you to build models of any size with google’s scalable infrastructure. It creates models using powerful TensorFlow Framework that controls many google products such as Google Photos and Google Cloud Speech. This service is associated with Google Cloud Dataflow, Google Cloud Storage, Google BigQuery and others.


Google is working with various technology partners to influence their own solutions with Cloud Machine Learning Engine. The Best example is SpringML which is using Cloud Machine Learning Engine to provide real-time analysis to its users.

Google hopes that it will help organizations to drive their business effectively and be successful with machine learning.

Popular posts

Leave a Reply

Your email address will not be published. Required fields are marked *