zlacker

The opposite way, the training images are there to support the model to generalize features.

Reproducing parts of existing images in the dataset is called overfitting and is considered a failure of the model.

>>realus+(OP)
how do you measure success?

i wrote an OCR program in college. we split the data set in half. you train it on one half then test it against the other half.

you can train stable diffusion on half the images, but then what? you use the image descriptions of the other half and measure how similar they are? in essence, attempting to reproduce exact replicas. but i guess even then it wouldn't be copyright if those images weren't used in the model. more like me describing something vividly to you and asking you to paint it and then getting angry at you because its too accurate

>>8n4vid+m4
FID score is a measure of success.

Instead of aiming to reproduce exact replicas, you use a classifier and retrieve the input of the last layer. Do it for both generated and original inputs, and then measure the differences in the statistics.

Wikipedia has a good article on this.

>>8n4vid+m4
You would not need have of the images to perform that test. No more than a handful of images to prove that the text representation will not produce a identical image to a given image that has had a description described.

They don't even produce the same image twice from the same description and a different random seed.