FastComposer: Tuning-Free Multi-subject Image Generation with Localized Attention

Abstract Diffusion models excel at text-to-image generation, especially in subject-driven generation for personalized images. However, existing methods are inefficient due to the subject-specific fine-tuning, which is computationally intensive and hampers efficient deployment. Moreover, existing methods struggle with multi-subject generation as they often blend identity among subjects. We present FastComposer which enables efficient, personalized, multi-subject text-to-image generation without fine-tuning. FastComposer uses subject embeddings extracted by an image encoder to augment the generic text conditioning in diffusion models, enabling personalized image generation based on subject images and textual…

• A
  Amazon
• M
  Microsoft
• MI
  Massachusetts Institute of Technology
• N
  Nvidia
• DS
  Defence Science and Technology Agency - Singapore

  Award: DST00OECI20300823
• DO
  Division of Computing and Communication Foundations

  Award: 1943349
• DO
  Division of Information and Intelligent Systems

  Award: 2105819
• MW
  MIT-IBM Watson AI Lab