Analogist: Out-of-the-box Visual In-Context Learning with Image Diffusion Model

1City University of Hong Kong,
2State Key Lab for Novel Software Technology, Nanjing University,
3Tianjin University
TL;DR: Analogist is a visual In-Context Learning approach that leverages pre-trained Diffusion Models for various tasks with few example pairs, without requiring fine-tuning or optimization.

Abstract

Visual In-Context Learning (ICL) has emerged as a promising research area due to its capability to accomplish various tasks with limited example pairs through analogical reasoning. However, training-based visual ICL has limitations in its ability to generalize to unseen tasks and requires the collection of a diverse task dataset. On the other hand, existing methods in the inference-based visual ICL category solely rely on textual prompts, which fail to capture fine-grained contextual information from given examples and can be time-consuming when converting from images to text prompts.

To address these challenges, we propose Analogist, a novel inference-based visual ICL approach that exploits both visual and textual prompting techniques using a text-to-image diffusion model pretrained for image inpainting. For visual prompting, we propose a self-attention cloning (SAC) method to guide the fine-grained structural-level analogy between image examples. For textual prompting, we leverage GPT-4V's visual reasoning capability to efficiently generate text prompts and introduce a cross-attention masking (CAM) operation to enhance the accuracy of semantic-level analogy guided by text prompts.

Our method is out-of-the-box and does not require fine-tuning or optimization. It is also generic and flexible, enabling a wide range of visual tasks to be performed in an in-context manner. Extensive experiments demonstrate the superiority of our method over existing approaches, both qualitatively and quantitatively.

Method

Overview of the proposed Analogist. A visual demonstration is defined by an example pair \(A\) (woman holding a cat) and \(A'\) (the same woman holding a tiger). Given a new image \(B\) (another cat), we format these three images into a \(2 \times 2\) grid and tackle this problem by fill the missing image via a pretrained Stable Diffusion inpainting model. The images are arranged into a \(2 \times 2\) grid and feed into a pretrained inpainting model. We employ GPT-4V to provide a proper text description (i.e., "close-up of a tiger's face") to further guide the inpainting process. During the process of model inference, Self-Attention Cloning (SAC) and Cross-Attention Masking (CAM) are introduced to encourage the model concentrate on the visual and textual prompts

Self-Attention Cloning (SAC)

The sub self-attention map \(\mathcal{M}_s(A',B')\) is set as the value of \(\mathcal{M}_s(A,B)\), denoting cloning the relation between \(A\) and \(B\) to that of \(A'\) and \(B'\).

Cross-Attention Masking (CAM)

The sub cross-attention map between text embedding and regions \(A\), \(A'\), and \(B\) are set to zero, making the semantic guidance more focused on region \(B'\).

Results

Image Colorization

Image Deblurring

Image Denoising

Low-light Enhancement

Image Editing

Image Translation

Style Transfer

Skeleton-to-image

Mask-to-image

Image Inpainting

More Applications

\(A\) and \(A'\) are aligned

\(A\) and \(B\) are aligned

BibTeX

@article{gu2024analogist,
      title     = {Analogist: Out-of-the-box Visual In-Context Learning with Image Diffusion Model},
      author    = {GU, Zheng and Yang, Shiyuan and Liao, Jing and Huo, Jing and Gao, Yang},
      journal   = {ACM Transactions on Graphics (TOG)},
      year      = {2024},
}