Search papers, labs, and topics across Lattice.
AttDiff-GAN, a novel hybrid framework, is introduced to improve facial attribute editing by combining GAN-based attribute manipulation with diffusion-based image generation. It addresses the inconsistency between one-step adversarial learning and multi-step diffusion denoising by decoupling attribute editing from image synthesis via feature-level adversarial learning. The framework also incorporates PriorMapper and RefineExtractor to enhance style-attribute alignment, leading to more accurate attribute editing and better preservation of non-target attributes on CelebA-HQ.
Achieve more precise facial attribute editing by decoupling attribute manipulation from image synthesis, sidestepping the optimization challenges of directly combining GANs and diffusion models.
Facial attribute editing aims to modify target attributes while preserving attribute-irrelevant content and overall image fidelity. Existing GAN-based methods provide favorable controllability, but often suffer from weak alignment between style codes and attribute semantics. Diffusion-based methods can synthesize highly realistic images; however, their editing precision is limited by the entanglement of semantic directions among different attributes. In this paper, we propose AttDiff-GAN, a hybrid framework that combines GAN-based attribute manipulation with diffusion-based image generation. A key challenge in such integration lies in the inconsistency between one-step adversarial learning and multi-step diffusion denoising, which makes effective optimization difficult. To address this issue, we decouple attribute editing from image synthesis by introducing a feature-level adversarial learning scheme to learn explicit attribute manipulation, and then using the manipulated features to guide the diffusion process for image generation, while also removing the reliance on semantic direction-based editing. Moreover, we enhance style-attribute alignment by introducing PriorMapper, which incorporates facial priors into style generation, and RefineExtractor, which captures global semantic relationships through a Transformer for more precise style extraction. Experimental results on CelebA-HQ show that the proposed method achieves more accurate facial attribute editing and better preservation of non-target attributes than state-of-the-art methods in both qualitative and quantitative evaluations.
