The Verge: Plenty of companies are taking advantage of machine learning to tag and search visual content. Pinterest lets you find visually-similar images in order to track down that recipe or jacket you’re looking for, and Pornhub is using machine learning to automatically identify porn stars in videos. Stock image company Shutterstock, though, has developed one of the more novel implementations of this sort of technology: using machine learning to identify the layout of images.
:format(webp)/cdn.vox-cdn.com/uploads/chorus_image/image/57113683/Screenshot_2017_10_12_14.03.59_preview.0.png)
1 comment:
I never imagined wanting to search for images with a certain layout, but I have to say that this identification and filtering software is definitely an interesting one. My favorite machine learning utility in this category, however, is the reverse Google image search function where you can drop an image you already have in Google, with or without keywords included, and it will spit back all the images that match that one or at least come close. The main issue with Google’s engine is that it only spits back those specific images and related images are harder to generate and find in this function. However, Shutterstock’s model could do exactly this if you knew what was in the photo and how it was laid out. I also like it for its graphic design utility and the ability to say where you want blank space compositionally. I feel like this search engine could come in handy for both business and personal use.
Post a Comment