Cloud

Box adds Google Cloud Vision integration, using machine learning to improve asset management

New image recognition capabilities within Box will allow users to index any text in a photo as metadata and more easily categorize their files.

With help from Google's machine learning-powered Cloud Vision API, Box users will now have access to more powerful image recognition and insights, the cloud storage firm announced in a Thursday blog post.

According to the post, "Images are the second most common and fastest growing content type in Box." Now, using machine learning through Google's API, Box customers will be able to better organize and manage their images, while also gleaning insights that can potentially make the assets more valuable.

The Cloud Vision API is the same technology that powers Google Photos. With the API's machine learning capabilities, Box will be able to recognize objects like flowers or landmarks in an image, the post said, and make them searchable by those terms. This could be especially helpful for inventory management in a retail company.

SEE: Big data in 2017: AI, machine learning, cloud, IoT, and more

One of the customer use cases that Box presented in its post was that of a media organization that uses the image recognition to automatically tag a large number of inbound images from freelance photographers. They previously had no way to tag each individual image, but now have new insights about the files they're managing.

What's even more useful for business customers is the ability to extract text from images that contain handwritten or typed words.

"Now, upon uploading images to Box, any objects and handwritten or typed text in images can be automatically detected and indexed as metadata, bringing actionable context to your content," the post said.

That means that images and screenshots of items such as legal contracts, research papers, books, and loan applications will all be indexed and searchable. This could have big implications for the finance and legal industries that often deal with massive amounts of paperwork.

A global real estate firm is using the optical character recognition (OCR) in this capability to digitize its process for lease agreements, the post said. This saves their employees time and money.

While the Cloud Vision API integrations marks one of Box's first forays into machine learning, the post mentioned that the company would be working to bring more machine learning features to users in the future.

The 3 big takeaways for TechRepublic readers

  1. Box is integrating Google's Cloud Vision API to offer better image recognition and metadata indexing to customers.
  2. The API will be able to recognize images like landmarks in a photo, while also indexing any written or typed text that appears in an image as well.
  3. Box is planning to bring more machine learning features in the future.

Also see

boximage.jpg
Image: Box

About Conner Forrest

Conner Forrest is a Senior Editor for TechRepublic. He covers enterprise technology and is interested in the convergence of tech and culture.

Editor's Picks

Free Newsletters, In your Inbox