arxiv:2410.22901

HelloMeme: Integrating Spatial Knitting Attentions to Embed High-Level and Fidelity-Rich Conditions in Diffusion Models

Published on Oct 30

· Submitted by

songkey on Nov 4

Upvote

Authors:

Shengkai Zhang ,

Abstract

We propose an effective method for inserting adapters into text-to-image foundation models, which enables the execution of complex downstream tasks while preserving the generalization ability of the base model. The core idea of this method is to optimize the attention mechanism related to 2D feature maps, which enhances the performance of the adapter. This approach was validated on the task of meme video generation and achieved significant results. We hope this work can provide insights for post-training tasks of large text-to-image models. Additionally, as this method demonstrates good compatibility with SD1.5 derivative models, it holds certain value for the open-source community. Therefore, we will release the related code (https://songkey.github.io/hellomeme).

View arXiv page View PDF Add to collection

Community

songkey

Paper author Paper submitter 12 days ago

•

edited 11 days ago

Abstract
We propose an effective method for inserting adapters into text-to-image foundation models, which enables the execution of complex downstream tasks while preserving the generalization ability of the base model. The core idea of this method is to optimize the attention mechanism related to 2D feature maps, which enhances the performance of the adapter. This approach was validated on the task of meme video generation and achieved significant results. We hope this work can provide insights for post-training tasks of large text-to-image models. Additionally, as this method demonstrates good compatibility with SD1.5 derivative models, it holds certain value for the open-source community.

Project Page | Code Page | Arxiv | ComfyUI