Bergur Thormundsson
Research expert covering technology in the Nordics, AI and smart cities
Get in touch with us nowError rates of large-scale visual recognition challenge 2010-2017
The statistic shows the best classification error rate achieved by computer vision algorithms tested on a large-scale visual recognition challenge, from 2010 to 2017. In 2015, the winning algorithm became the first to surpass the average human classification error rate of five percent, and by 2017 machine learning algorithms were able to achieve a classification error rate of 2.3 percent, making fewer than half the number of classification errors as a human.
