Style Ambiguity Loss Using CLIP

James Baker
10/2/2024
cs.CV

Official code: jamesBaker361/clipcreate (1 star, last commit 2025-06-06).

Abstract

In this work, we explore using the style ambiguity training objective, originally used to approximate creativity, on a diffusion model. However, this objective requires the use of a pretrained classifier and a labeled dataset. We introduce new forms of style ambiguity loss that do not require training a new classifier or a labeled dataset. Instead of using a classifier, we generate centroids in the CLIP embedding space, and images are classified based on their relative distance to said centroids. We find the centroids via K-means clustering of an unlabeled dataset, as well as using text labels to generate CLIP embeddings, to be used as centroids. Code is available at https://github.com/jamesBaker361/clipcreate

View on arXivView PDF

Code Implementations(1)

jamesBaker361/clipcreateOfficial100%
10PythonDec 8, 20231 years ago

Cite this paper

@article{baker2024style,
  title  = {Style Ambiguity Loss Using CLIP},
  author = {James Baker},
  year   = {2024},
  eprint = {2410.02055},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2410.02055v3}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2410.02055.svg)](https://codeofpaper.com/paper/2410.02055)

Discussion