DreamSteerer: Enhancing Src Img Conditioned Editability using Personalized Diff Models (Yu et al.)
About this lesson
DreamSteerer: Enhancing Source Image Conditioned Editability using Personalized Diffusion Models Zhengyang Yu, Zhaoyuan Yang and Jing Zhang In Advances in Neural Processing Systems (NeurIPS), 2024. ABSTRACT Recent text-to-image (T2I) personalization methods have shown great premise in teaching a diffusion model user-specified concepts given a few images for reusing the acquired concepts in a novel context. With massive efforts being dedicated to personalized generation, a promising extension is personalized editing, namely to edit an image using personalized concepts, which can provide more precise guidance signal than traditional textual guidance. To address this, one straightforward solution is to incorporate a personalized diffusion model with a text-driven editing framework. However, such solution often shows unsatisfactory editability on the source image. To address this, we propose DreamSteerer, a plug-in method for augmenting existing T2I personalization methods. Specifically, we enhance the source image conditioned editability of a personalized diffusion model via a novel Editability Driven Score Distillation (EDSD) objective. Moreover, we identify a mode trapping issue with EDSD, and propose a mode shifting regularization with spatial feature guided sampling to avoid such issue. We further employ two key modifications on the Delta Denoising Score framework that enable high-fidelity local editing with personalized concepts. Extensive experiments validate that DreamSteerer can significantly improve the editability of several T2I personalization baselines while being computationally efficient. PAPER https://neurips.cc/virtual/2024/poster/94953
DeepCamp AI