Training-Free Identity Preservation in Stylized Image Generation Using Diffusion Models
By: Mohammad Ali Rezaei , Helia Hajikazem , Saeed Khanehgir and more
Potential Business Impact:
Keeps faces the same when changing picture styles.
While diffusion models have demonstrated remarkable generative capabilities, existing style transfer techniques often struggle to maintain identity while achieving high-quality stylization. This limitation is particularly acute for images where faces are small or exhibit significant camera-to-face distances, frequently leading to inadequate identity preservation. To address this, we introduce a novel, training-free framework for identity-preserved stylized image synthesis using diffusion models. Key contributions include: (1) the "Mosaic Restored Content Image" technique, significantly enhancing identity retention, especially in complex scenes; and (2) a training-free content consistency loss that enhances the preservation of fine-grained content details by directing more attention to the original image during stylization. Our experiments reveal that the proposed approach substantially surpasses the baseline model in concurrently maintaining high stylistic fidelity and robust identity integrity, particularly under conditions of small facial regions or significant camera-to-face distances, all without necessitating model retraining or fine-tuning.
Similar Papers
Data Synthesis with Diverse Styles for Face Recognition via 3DMM-Guided Diffusion
CV and Pattern Recognition
Creates fake faces that fool face recognition.
FastFace: Tuning Identity Preservation in Distilled Diffusion via Guidance and Attention
CV and Pattern Recognition
Makes AI art generators create faces faster.
IP-FaceDiff: Identity-Preserving Facial Video Editing with Diffusion
CV and Pattern Recognition
Changes faces in videos with text commands.