Could anyone kindly point me to revenant literature in computer vision/graphics that detail on metrics for image/visual complexity and how to regularize that?

More specifically, I am trying to design a neural algorithm for creating business logos and one major difference from image generation problems in CV/CG is the simplicity required of logos. I have been thinking along the lines of adding a visual/image complexity term to be regularized and penalized as the algorithm learns the balance between lower-level style and higher-level meaning/content of multidimensional input images and descriptions. I have been searching along but couldn't find related resources... Could anyone shed some light on that?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.