Over at Perficient Digital, an image recognition study was released that looked at the accuracy of the respective technologies. The study found that Google Vision beat out the competition, which included Amazon AWS Rekognition, IBM Watson and Microsoft Azure Computer Vision.
As you can see below, Google has scored the highest among the technologies:
There were two humans who collected and tagged 2,000 images in four different categories:
Each of these categories had about 500 images. Images were collected and tagged between November 30, 2018 and January 8, 2019. Both humans came up with and assigned five tags to describe each image. Perficient ran each of the 2,000 images through each of the image analysis APIs listed above and looked at the results where a unique set of labels/tags for each image from each API. After every tag for the image was assigned a value, the next image was presented. This process took place between April 12 and May 9.
The study really dug into multiple ways to slice and dice the data. You can check out the whole study over here.