Facebook photo search set to get way more visual
With popular networking site's latest search tool, you can easily recognise what's in an image
A man is silhouetted against a video screen with an Facebook logo as he poses with an Samsung S4 smartphone in this photo illustration taken in the central Bosnian town of Zenica, August 14, 2013. PHOTO: REUTERS
Facebook's search option may soon get much more advanced as the popular social networking site is developing a feature that would be able to recognise what’s in an image and let you search images based on their content irrespective of the fact where it was tagged properly or not.
According to a blog post by Director of Applied Machine Learning Joaquin Candela, "Until recently, online search has always been a text-driven technology, even when searching through images."
"Whether an image was discoverable was dependent on whether it was sufficiently tagged or had the right caption – until now," it added.
Facebook vows aggressive investment as earnings surge
With this tool, for instance, you can recognise what's in an image, what type of scene it is, if it's a well-known landmark, and so on.
These tools were originally envisioned to help the visually impaired navigate the service, discerning what's in a photo just by scanning it. But today's news shows general Facebook users have a lot to benefit from the feature as well.
Facebook's general-purpose AI platform, FBLearner Flow, is currently running 1.2 million AI experiments a month – six times more than it was just a year ago.
Asia is now Facebook’s biggest region
Built on top of that, Facebook is using Lumos, a specialised platform for image and video understanding which can help identify features in images and video automatically. For users, that capability will help pinpoint searches to the exact pic they're looking, for instance, when you search your old photos, you'll be able to look for images where you're wearing a black shirt or red dress, or where the people in the image are dancing.
For Facebook, this automation will make it easier to identify inappropriate content and spam.
According to a blog post by Director of Applied Machine Learning Joaquin Candela, "Until recently, online search has always been a text-driven technology, even when searching through images."
"Whether an image was discoverable was dependent on whether it was sufficiently tagged or had the right caption – until now," it added.
Facebook vows aggressive investment as earnings surge
With this tool, for instance, you can recognise what's in an image, what type of scene it is, if it's a well-known landmark, and so on.
These tools were originally envisioned to help the visually impaired navigate the service, discerning what's in a photo just by scanning it. But today's news shows general Facebook users have a lot to benefit from the feature as well.
Facebook's general-purpose AI platform, FBLearner Flow, is currently running 1.2 million AI experiments a month – six times more than it was just a year ago.
Asia is now Facebook’s biggest region
Built on top of that, Facebook is using Lumos, a specialised platform for image and video understanding which can help identify features in images and video automatically. For users, that capability will help pinpoint searches to the exact pic they're looking, for instance, when you search your old photos, you'll be able to look for images where you're wearing a black shirt or red dress, or where the people in the image are dancing.
For Facebook, this automation will make it easier to identify inappropriate content and spam.