Latest in Gear

Image credit: Thomas Trutschel via Getty Images

Facebook trained image recognition AI with billions of Instagram pics

The method could allow models to be trained with larger data sets and less human involvement.
358 Shares
Share
Tweet
Share
Save

Sponsored Links

Thomas Trutschel via Getty Images

Training deep learning models to recognize images, as well as objects within those images, takes quite a bit of effort. Often, each training image has to be labeled by humans and when you're using millions of images, that process becomes rather labor-intensive. Scaling up to billions of images becomes nearly impossible. So, Facebook has been working on a way to train deep learning models with limited human supervision. Instead, its researchers have turned to public images that are, in a way, already labeled -- with hashtags.

With this method, Facebook researchers and engineers trained image recognition networks with up to 3.5 billion Instagram images labeled with as many as 17,000 hashtags. After training the computer vision system with one billion images and 1,500 hashtags, it was able to achieve 85.4 percent image recognition accuracy on the popular benchmarking tool ImageNet. It beat out the previous state-of-the-art model, which achieved an accuracy rating of 83.1 percent.

The work shows that weakly supervised training is a valid option moving forward, opening up deep learning model training to larger data sets and, possibly, more accurate image recognition and classification. Better image recognition could improve AI-generated audio captions of photos for the visually impaired, but Facebook says there are other useful applications as well. Using hashtags as labels for computer vision could impact how Facebook ranks images in feeds and improve an AI system's understanding of video footage.

"As training data sets get larger, the need for weakly supervised -- and, in the longer term, unsupervised -- learning will become increasingly vital," said Facebook. "Understanding how to offset the disadvantages of noisier, less curated labels is critical to building and using larger-scale training sets."

Click here to catch up on the latest news from F8 2018!

All products recommended by Engadget are selected by our editorial team, independent of our parent company. Some of our stories include affiliate links. If you buy something through one of these links, we may earn an affiliate commission.
Comment
Comments
Share
358 Shares
Share
Tweet
Share
Save

Popular on Engadget

Engadget's 2019 Back-to-School Guide

Engadget's 2019 Back-to-School Guide

View
Apple warns against storing its titanium credit card in leather

Apple warns against storing its titanium credit card in leather

View
Google's next Nest Mini speaker could be wall-mountable

Google's next Nest Mini speaker could be wall-mountable

View
Microsoft tests more control for apps that restart with Windows 10

Microsoft tests more control for apps that restart with Windows 10

View
Terminator T-800 and The Joker are coming to 'Mortal Kombat 11'

Terminator T-800 and The Joker are coming to 'Mortal Kombat 11'

View

From around the web

Page 1Page 1ear iconeye iconFill 23text filevr