p Over the past few years, Facebook’s made significant strides with their image recognition AI, helping to power a more advanced version of their on-platform image search tool - which probably hasn’t received as much attention as it deserves.

There are a heap of ways you can use image recognition to advance your social marketing strategy – for example, if you can identify people who regularly share images of products similar or related to your own on Facebook, and within your local region, that could present an opportunity for you to reach out and offer a discount or special offer to them in the hopes of establishing a connection and getting them to share the word about your business with their network.

It's not as simple a process as blanketing groups with ads, but getting a selected users to post a personal update is more conducive to viral sharing, and targeting those who are already sharing related images could be a great way to facilitate such.

With Facebook’s search capacity, this is already possible, and this week, The Social Network has shared a new research paper which underlines just how advanced and focused their image search tools have already become.

First off, back in February Facebook released a research update on their advance image recognition AI and how it was now able to identify more actions and objects within photos, which have been built into their on-platform search function.

As you can see in these examples, you can now search for places or events and Facebook will be able to identify such, even if the image wasn’t tagged with that information.

The text above is a summary, you can read full article here.