Google's Cloud Vision AI - Now We Can All Play
Written by David Conrad   
Monday, 22 February 2016

Google announced Cloud Vision at the end of 2015, but it was in limited preview until now. Now, not only can we get our hands on the beta, we have inital pricing to work out if it is worth it. 


The Cloud Vision API offers Google's computer vision system to anyone who wants to use it and there is a free tier. 

The list of processing options given is:

  • Insights from your images: Cloud Vision API detects broad sets of objects in your images — from flowers to popular landmarks
  • Inappropriate content detection: Powered by Google SafeSearch, Cloud Vision API moderates content from your crowd sourced images by detecting different types of inappropriate content.
  • Image sentiment analysis: Cloud Vision API can analyze emotional attributes of people in your images, like joy, sorrow and anger, along with detecting popular product logos.
  • Text extraction: Optical Character Recognition (OCR) enables you to detect text within your images, along with automatic language identification across a broad set of languages.

 As noted previously face detection is supported but not face recognition - to avoid privacy problems. 




Of course what really matters is the price. The service is free until March 1st and after that charges will apply. The following rates are for up to 20 million uses per month; beyond that, you will have to talk to Google: 

FEATURE    1 - 1K 1K - 1M  1 - 5M  5 - 20 M 
Optical Character Recognition (OCR) Free $2.50 $2.00 $0.60
Detection of Image Attributes, Landmark, Facial, Logo, or Explicit Content Free $2.50 $2.00 $0.60
Label Detection Free $5.00 $4.00 $2.00


The free tier of 1000 uses per month seems like a good way of finding out or experimenting. It could even be useful in education.

The documentation points out that as a beta the API is subject to change and that it isn't intended for real time use in critical applications. So don't try to make use of it for robot guidance, for example.

To use the API you have to enable it in the developer console and provide a way to pay for what you might use. The simplest way of using the API is to send an image, Base64 encoded in a JSON structure that carries the data and what you want to do to it. Send the JSON to the endpoint and you get JSON back carrying the results. You can also pass a url to an image stored in Google Cloud Storage rather than having to upload the image each time.  



More Information

Cloud Vision API

Related Articles

Google Gears Up To Sell AI Vision 

TensorFlow - Googles Open Source AI And Computation Engine

Emotion Detection Using Project Oxford

RankBrain - AI Comes To Google Search

Project Oxford Extends Reach

The Flaw In Every Neural Network Just Got A Little Worse

Google Files AI Patents

Facebook Shares Deep Learning Tools



To be informed about new articles on I Programmer, sign up for our weekly newsletter, subscribe to the RSS feed and follow us on, Twitter, FacebookGoogle+ or Linkedin




Next.js Optimizes Packet Imports

Next.js has been updated with faster server startups, lower memory use and optimized packet imports.

Learn About RISC-V From Linux Foundation on edX

The Linux Foundation has a new free course on the edX platform for those who would like to learn about, and experiment with, RISC-V. A verified track, to earn a certificate of completion, is also [ ... ]

More News


Summer SALE Kindle 9.99 Paperback $10 off!!




or email your comment to:

Last Updated ( Monday, 27 September 2021 )