Since photos and videos are the most popular and high-engagement posts on the social networking website, it comes as no surprise that Facebook keeps developing its tech in that area.
The California-based tech giant has enhanced its artificial intelligence tech — Lumos — which can now search photos of you without you being even tagged in them because it can ‘recognise what’s in an image; at the pixel level’.
Lumos will now recognise everything in the image, what kind of a scene it is as well as any landmark in the image.
“Whether an image was discoverable was dependent on whether it was sufficiently tagged or had the right caption — until now.That’s changing because we’ve pushed computer vision to the next stage with the goal of understanding images at the pixel level,” writes Joaquin Quinonero Candela, Director of Applied Machine Learning, Facebook.
With the new update, Lumos will be able to aptly search for an image even if it doesn’t have a text caption — which more than often isn’t relevant to an uploaded photograph anyways.
“This, in turn, helps us better describe photos for the visually impaired and provide better search results for posts with images and videos,” he adds.
The AI was mainly used to support the visually-impaired using the social media network using the alt-text which will now also be able to track and recognise 12 new actions like riding a horse or people walking, dancing, or playing an instrument.
“With computer vision models getting pixel perfect and Facebook advancing into video and other immersive formats, Lumos will help unlock new possibilities in a reliable, fast, and scalable way and pave the road for richer product experiences in the near future,” Candela concluded.
Artificial intelligence tech is emerging as the next big thing among other techs in the world of internet and since this is just ‘scratching the surface’, expect Lumos, as well as the social networking platform to get better and more lively in the time to come.