Tuning-Free Image Customization with Image and Text Guidance
This provides an efficient solution for practical applications like image synthesis and design, though it is incremental as it builds on existing diffusion model techniques.
The paper tackles the problem of unintended changes and time-consuming fine-tuning in image customization with diffusion models by introducing a tuning-free framework that uses both text and image guidance to edit specific regions within seconds, outperforming previous methods in evaluations.
Despite significant advancements in image customization with diffusion models, current methods still have several limitations: 1) unintended changes in non-target areas when regenerating the entire image; 2) guidance solely by a reference image or text descriptions; and 3) time-consuming fine-tuning, which limits their practical application. In response, we introduce a tuning-free framework for simultaneous text-image-guided image customization, enabling precise editing of specific image regions within seconds. Our approach preserves the semantic features of the reference image subject while allowing modification of detailed attributes based on text descriptions. To achieve this, we propose an innovative attention blending strategy that blends self-attention features in the UNet decoder during the denoising process. To our knowledge, this is the first tuning-free method that concurrently utilizes text and image guidance for image customization in specific regions. Our approach outperforms previous methods in both human and quantitative evaluations, providing an efficient solution for various practical applications, such as image synthesis, design, and creative photography.