用CLIP无监督生成风格中心点，提升扩散模型的创意多样性。
Style Ambiguity Loss Using CLIP
- 通过CLIP嵌入空间聚类生成风格中心点，无需训练分类器
- 在无标签数据上实现风格模糊性损失，使生成结果更富创意
- 适合希望提升生成艺术风格多样性的研究人员
本文探索将原本用于近似创造力的风格模糊性训练目标应用于扩散模型。该目标传统上依赖预训练分类器和标注数据集，但本研究提出新形式的风格模糊性损失，无需训练新分类器或使用标注数据。我们通过K-means聚类无标签数据，并利用文本标签生成的CLIP嵌入作为聚类中心，以图像与中心点的相对距离进行风格分类。实验表明，该方法有效提升了生成结果的风格多样性与创造性。代码已开源：https://github.com/jamesBaker361/clipcreate。
原文摘要 · Abstract (English)
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
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。