The Facebook image search system now sports substantially more power, thanks to the social site’s artificial intelligence team, for better search results…
The artificial intelligence team at Facebook just built a visual search system that is able to recognize content appearing in pictures to return relevant user search results. This provides users with more search power to find or rediscover content.
Facebook Image Search Improved Significantly
“Think about the last post you liked — it most likely involved a photo or video. But, 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,” Joaquin Quiñonero Candela wrote.
The visual search system, named “Lumos,” was initially created by the social platform to serve its visually impaired users. This helped those users to understand photo content. Facebook eventually realized the technology is useful to all its users.
The Facebook image search technology is able to detect and recognize objects, scenes, places, clothing, and animals in both still images and videos. Lumos performs these complex tasks with “cutting-edge deep learning techniques to process billions of photos and understand their semantic meaning.”
The image search system even works on photos without tags. So, searching for “blue suede shoes photo” will return query results relevant to the search term. “We’ve built a search system that leverages image understanding to sort through this vast amount of information and surface the most relevant photos quickly and easily,” the company explains.
Facebook is changing its News Feed post ranking algorithm to reduce spam, misleading titles, and clickbait content. Additionally, the Facebook Trending stories section is also undergoing three key changes.
Thanks for taking a few moments of your time to read this; if you like this information, please share it with your friends!