Hi everyone!
A few weeks ago OpenAI (the creators of GPT-3) released this new amazing model called CLIP. CLIP is a model that computes the similarity between text and images, and as it has been trained on a huge dataset, it generalizes to a lot of different domains. This means that you can use it out of the box to solve a lot of computer vision problems.
However, I couldn’t find an easy way to test its capabilities, so I decided to create a simple streamlit app myself. I wanted to give everyone (including myself) the opportunity to test what CLIP is capable of from the comfort of the browser.
That’s why I created www.clipplaygroud.co!
After playing with it for a while I am impressed! You can use CLIP for a huge variety of problems. I give you some examples in the app on how to use it, but I am sure you will come up with other cool applications!
If you are interested in learning more about how I built it and deploy it check out this thread
Try it and let me know what you think!