{
  "id": "2311.10093",
  "version": "v1",
  "published": "2023-11-16T18:59:51.000Z",
  "updated": "2023-11-16T18:59:51.000Z",
  "title": "The Chosen One: Consistent Characters in Text-to-Image Diffusion Models",
  "authors": [
    "Omri Avrahami",
    "Amir Hertz",
    "Yael Vinker",
    "Moab Arar",
    "Shlomi Fruchter",
    "Ohad Fried",
    "Daniel Cohen-Or",
    "Dani Lischinski"
  ],
  "comment": "Project page is available at https://omriavrahami.com/the-chosen-one",
  "categories": [
    "cs.CV",
    "cs.GR",
    "cs.LG"
  ],
  "abstract": "Recent advances in text-to-image generation models have unlocked vast potential for visual creativity. However, these models struggle with generation of consistent characters, a crucial aspect for numerous real-world applications such as story visualization, game development asset design, advertising, and more. Current methods typically rely on multiple pre-existing images of the target character or involve labor-intensive manual processes. In this work, we propose a fully automated solution for consistent character generation, with the sole input being a text prompt. We introduce an iterative procedure that, at each stage, identifies a coherent set of images sharing a similar identity and extracts a more consistent identity from this set. Our quantitative analysis demonstrates that our method strikes a better balance between prompt alignment and identity consistency compared to the baseline methods, and these findings are reinforced by a user study. To conclude, we showcase several practical applications of our approach. Project page is available at https://omriavrahami.com/the-chosen-one",
  "revisions": [
    {
      "version": "v1",
      "updated": "2023-11-16T18:59:51.000Z"
    }
  ],
  "analyses": {
    "keywords": [
      "text-to-image diffusion models",
      "game development asset design",
      "consistent character generation",
      "text-to-image generation models",
      "prompt alignment"
    ],
    "note": {
      "typesetting": "TeX",
      "pages": 0,
      "language": "en",
      "license": "arXiv",
      "status": "editable"
    }
  }
}