I am learning CLIP and I read the following information:
arxiv
openai/clip-vit-large-patch14
huggingface-clip-api

I want to know more about how image encoder and text encoder are implemented.
How are they ultimately compared in detailed?
Are there any recommended study materials?

I explored the provided resources to understand the implementation details of the image encoder and text encoder in CLIP. I expected to find a comprehensive explanation of their architectures, how they process inputs, and the specific techniques used to compare their outputs. While I did find some high-level overviews, I was looking for a more detailed, step-by-step explanation of the implementation and comparison process.