Facebook’s Lumos AI Tool Will Help You In Finding Untagged Images Easily


Facebook is working on a new photo identification feature using image search system named - Lumos AI. Through this feature, it will be easier for users to search for a specific image even if it is not properly tagged or captioned.

This is not the first time when Facebook is working on such a feature. Last year, the social networking giant introduced a similar AI based image search system for helping visually challenged users in identifying image elements. Now, Facebook is taking it one level up by integrating it into image search for regular users. The AI based system performs image searches by identifying elements in the picture, for example - the color of clothes, background, objects, place, and more.

Apple is the World’s Most Valuable Brand According to New Survey

Google has also tried a similar system in its Google Photos app, which finds images in the library on the basis of what is in the photo such as object, location, and more. Facebook's Lumos AI technology will make it easier for users to search via image content without the need for remembering the exact text caption used for it. You can base your image search on any elements of the picture such as the place where it was shot or the color of T-shirt that you were wearing or if there was an animal in the picture. The AI based search will identify the elements you searched for and bring you the correct results. It will surely save a lot of time.

Lumos AI Built On FBLearner Flow Machine Learning

In a blog post, Facebook's Director of Applied Machine Learning Joaquin Quinonero Candela said:

"We've pushed computer vision to the next stage with the goal of understanding images at the pixel level. This helps our systems do things like recognize what's in an image, what type of scene it is, if it's a well-known landmark, and so on. This, in turn, helps us better describe photos for the visually impaired and provide better search results for posts with images and videos."

To save its engineers from the pain of scaling machine learning pipelines to tackle real-time traffic on the website, Facebook developed a new general-purpose platform named as FBLearner Flow that applies the new AI tool on a wider basis. Facebook is currently conducting 1.2 million AI experiments on FBLearner Flow. And Lumos AI method was built on top of FBLearner for image and video content identification.

Commenting on FBLearner Flow, Candela said:

"As this platform has become more widely used, we've continued to build on top of it. From tools to automate the process of machine learning to dedicated content understanding engines, we've built an active ecosystem that allows engineers to write training pipelines that parallelise over many machines so it can be reused by any engineer at the company."

Regarding the rollout of this new image search feature, Facebook has not yet revealed a specific timeline. However, it will initially be available for users in the US.