Good publication to review material:
CognArtive: Large Language Models for Automating Art Analysis and Decoding Aesthetic Elements at https://arxiv.org/html/2502.04353v1
And the wonderful visualization to analyze the data over the years that was derived from WikiArt.
Autocritic: an open-source system that uses classic art theory to evaluate images and steer generative systems. It distills historical works by Wölfflin, Kandinsky, Arnheim, Klee, and others into structured “critic cards” that teach an LLM how to see through each theorist’s lens.
https://github.com/mccoyspace/autocritic
Art analyzer:
This is an app that uses GPT vision to provide the critique. Users can upload an imagen and get a critique. More LLM based
https://github.com/ericblue/art-analyzer
Art Critique and Creation tool
https://github.com/SamaMostafa03/Art-Critique-and-Creation-Tool
It has a critque mode which Analyzes an artist’s work and provides a critique of various elements, such as color palettes, composition, texture, and style, which can help artists improve or refine their works
Older one : MLArt-critc
This is based on traditional ML learning methods on a Microsoft framework in 2018 and 2019 with a movie to make it easy to implement: https://www.youtube.com/watch?v=-C8bm5kGDTI
