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

By Guangxuan Xiao, Tianwei Yin, William T. Freeman, Frédo Durand, Song Han. International Journal of Computer Vision, 2023

405 citations recorded by the source corpus.

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 instructions with only forward passes. To address the identity blending problem in the multi-subject generation, FastComposer proposes cross-attention localization supervision during training, enforcing the attention of reference subjects localized to the correct regions in the target images. Naively conditioning on subject embeddings results in subject overfitting. FastComposer proposes delayed subject conditioning in the denoising step to maintain both identity and editability in subject-driven image generation. FastComposer generates images of multiple unseen individuals with different styles, actions, and contexts. It achieves 300×\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\times $$\end{document}–2500×\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\times $$\end{document} speedup compared to fine-tuning-based methods and requires zero extra storage for new subjects. FastComposer paves the way for efficient, personalized, and high-quality multi-subject image creation. Code, model, and dataset are available here (https://github.com/mit-han-lab/fastcomposer).

Subfields: Computer Graphics and Computer-Aided Design, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/1a3d6119d9513ad27fa4fc3262e517ec6a6d2261

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Graphics and Computer-Aided Design

In Computer Vision and Pattern Recognition

Citations in the corpus

OminiControl: Minimal and Universal Control for Diffusion Transformer (2024, neutral): “To address this limitation, some works have explored image-based control methods [2, 14, 15, 18, 26– 30, 37, 42, 46, 50, 52, 54, 56, 57, 57, 60, 61, 63], enabling users to specify their intentions through reference images or visual hints, offering more precise guidance than text alone [44].”
AnyDoor: Zero-shot Object-level Image Customization (2023, neutral): “Fastcomposer [52] binds the image representation with certain text embeddings to do multiple-person generation.”
OmniGen: Unified Image Generation (2024, neutral): “Furthermore, OmniGen jointly models text and images within a single model, rather than independently modeling different input conditions with separate encoders as in existing works [8, 64, 68, 70, 72] which lacks interaction between different modality conditions.”
InstantBooth: Personalized Text-to-Image Generation without Test-Time Finetuning (2023, methodology): “The second is encoder-based methods [6, 15, 19, 22, 38, 39, 41], which learn image embeddings from the input images as the representation of the concept.”