Object detection for web components on page from library of 300+ components

[agustinagg]

Hello, I am looking for suggestions or recommendations for how to build an object detection model to identify what web components (out of a library of 300+ components) are present on a page. I am planning to feed the model a full page screenshot of that webpage and ideally it would identify each component on the page and provide a match from the component library. 

Has anyone attempted something like this before? Or trained a model on webpages before? I am curious to know how AutoML performs on dense webpages with both text and images.

I would greatly appreciate any guidance or comments on how to approach solving this problem. I am new to AutoML vision and am not even sure this is the right tool for my needs.

Many thanks,

Agustina