Skip to main content
ImageScan
L
Written by Lee Buchanan
Updated over 6 years ago

At Miappi we don’t just collect and display content for our customers, we analyze it in detail to reveal the most compelling stories...and now we can do it faster, much faster in fact.

A human moderator can review and consider about 2000 Tweets per day. Our technology can review 200,000 Tweets per day. That’s quite a helping hand when time’s tight and you need the right content, fast!

ImageScan is the latest Miappi feature designed to make it even easier to find and display your most valuable earned and owned content. ImageScan automatically searches all incoming images, scanning them for must-show (or must-hide!) content. If you don’t want to spend time moderating your content manually, post-by-post, ImageScan will do the heavy-lifting for you, sifting through your incoming data to uncover the best stories.

ImageScan lets you understand the content of an image by using powerful machine learning models. It quickly classifies images into thousands of categories (e.g., "sailboat", "lion", "Eiffel Tower"), detects individual objects and faces within images, and finds and reads printed words contained within images.

You can build metadata on your image catalogue, moderate offensive content, or enable new marketing scenarios through image sentiment analysis.

On request, we can apply one or more the following image filters:

  • Faces (Only flag images containing faces)

  • Minimum face count (Only flag images with e.g. more than two faces)

  • Labels (Generic detection of a 'label' e.g. boat, plane, tree, car etc.)

  • Landmarks (Waterfall, mountain, beach, Eiffel Tower etc).

  • Optical character recognition (Read words within an image)

  • Web detection (Like labels but specific to an identifiable person, place or event).

  • Adult/nudity (Porn usually!)

Using this powerful image analysis feature we have given you the ability to instantly showcase the best content on your Miappi social wall with the least amount of effort.

Did this answer your question?