MONKEY: Masking ON KEY-Value Activation Adapter for Personalization
This work addresses a specific issue in image generation personalization for users seeking more control over subject and background integration, representing an incremental improvement over existing methods.
The paper tackles the problem of personalizing diffusion models to generate images that accurately incorporate a given subject while adhering to text prompts, by using automatically generated masks to restrict subject tokens and allow text to control the background, resulting in high prompt and source image alignment as validated by a user study.
Personalizing diffusion models allows users to generate new images that incorporate a given subject, allowing more control than a text prompt. These models often suffer somewhat when they end up just recreating the subject image and ignoring the text prompt. We observe that one popular method for personalization, IP-Adapter, automatically generates masks that segment the subject from the background during inference. We propose to use this automatically generated mask on a second pass to mask the image tokens, thus restricting them to the subject, not the background, allowing the text prompt to attend to the rest of the image. For text prompts describing locations and places, this produces images that accurately depict the subject while definitively matching the prompt. We compare our method to a few other test time personalization methods, and find our method displays high prompt and source image alignment. We also perform a user study to validate whether end users would appreciate our method. Code available at https://github.com/jamesBaker361/monkey