Want to create an API with Python that can be used to manually create ML training dataset

I've been working on a project for some time now where I'm trying to classify a bunch of images using a software built for mapping with the front end in js and the back end in python. I'm running into a lot of problems with it, so I want to transition to something else and start over. Basically, I have a lot of images in a directory. I want my app to show them one at a time, receive user input, record that user input, then show the next image. The end goal is to use it to create a training dataset manually.

I want it to be as simple as possible but all the frameworks I'm looking into are overcomplicated, and I'm lost on where to start. I would appreciate any suggestions