The folks over at Perficient Digital released an image recognition study looking at the accuracy of the respective technologies. The study found that Google Vision beat out the competitors including Amazon AWS Rekognition, IBM Watson and Microsoft Azure Computer Vision.
The score. Here is the overall scoring, you can see how Google scored the highest amongst the technologies:
The methodology. Two humans collected and tagged 2,000 images in four distinct categories including people, landscapes, charts and products. Each category has approximately 500 images. Images were collected and tagged from November 30, 2018 through January 8, 2019. Each of the humans came up with and assigned five tags to describe each image. Perficient ran all 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. When each and every tag for the image was assigned a value, the next image was presented. This ranking process took place from April 12 to May 9.
A lot more here. The study really dug into multiple ways to slice and dice the data, so we recommend you review the whole study over here.
Why we care. First is that image recognition technology is pretty accurate and is getting better every day. With a confidence score of 80% the study found that the scores for ‘human hand tagged’ is basically equivalent to the results for Amazon AWS Rekognition, Google Vision, and Microsoft Azure Computer Vision.
Second, we should note that while Google and Bing both do image recognition in search – you would assume that what each search engine is using is probably a little more advanced and state-of-the-art than what the companies release to the public with the APIs. So you would have to assume image search in Google and Bing are even better when it comes to image recognition.
About The Author