Compiling...
From 2010 to 2017, the AI community hosted a massive annual computer vision competition called the ImageNet Challenge (ILSVRC). Researchers were tasked with building an AI that could correctly classify 1.2 million images into 1,000 different categories.
The architectures that won this competition were so effective that they became industry standards. Today, instead of guessing how many layers or filters to use, developers simply download these pre-built, proven architectures.
Let's look at the three most influential CNN architectures in history.
