Skip to main content

Experience Management

Google Releases Cloud Vision API. I Eat Crow.


Last year Google introduced groundbreaking image recognition technology as part of their release of Google Photos.  Deep neural networks could identify objects, people and places inside of photographs with astonishing accuracy, completely autonomously.
As fascinating as this technology is, I was certain that Google would keep it locked away inside their own products like Google Photos and Google Drive for the foreseeable future.  I did not anticipate it being sold standalone as an API.
I was wrong.  Two weeks ago, Google announced a beta version of their Cloud Vision API, allowing anyone to process images and identify things like everyday objects, faces (w/sentiment) and text or logos.
For example, processing the picture below with the Cloud Vision API that identifies common objects …  
faulkner
yields the following results:

“description”: “dog”, “score”: 0.89208293

The future is now.
The price is right, too.  The API is free for the first 1000 images per month (great for experimenting with new ideas) and then costs a fraction of a penny per call.  The unit price drops if you process more than 1,000,000 or 5,000,000 images per month.  Processing text or objects in one million images per month would cost a few thousand dollars.  Having priced similar enterprise-class software in the past, this is a bargain.  Most OCR or image recognition companies won’t even give you the time of day for this price.
We are particularly interested in technology like the Cloud Vision API for enriching content with metadata and making it easier to find in enterprise search applications.  Content is being created faster than it can be manually tagged and classified, making it difficult to find.  Try finding a picture from a few months ago in your cellphone’s camera roll.  Being able to search for text or objects inside the images makes this much easier.  Imagine letting your customers reorder a product by uploading a picture of the empty box or packaging.  Or automatically classifying similar items for a “related products” promotion.
I am very excited to have the Google Cloud Vision API available to us for photo enrichment in enterprise search application.  If you would like to learn more or have a project in mind, contact us at GooglePractice@perficient.com.
 
 

Leave a Reply

Your email address will not be published.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

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Chad Johnson

Chad is a Principal of Search and Knowledge Discovery at Perficient. He was previously the Director of Perficient's national Google for Work practice.

More from this Author

Follow Us