Why is the ImageNet Large Scale Visual Recognition Challenge considered the World Cup of computer vision?

The answer

The ImageNet Large Scale Visual Recognition Challenge (ILSVRC), launched in 2010, tasked teams with classifying 1.2 million images into 1,000 categories with the lowest error rate. In its first year, the best top-5 error rate was about 28%. By 2012, AlexNet slashed it to 15.3%, and by 2015, ResNet achieved 3.57% - surpassing the estimated 5.1% human error rate on the same task. The competition ran annually until 2017 and served as the single most important benchmark driving advances in deep learning for computer vision.