HairDiffusion: Vivid Multi-Colored Hair Editing via Latent Diffusion

TL;DR

HairDiffusion introduces a novel diffusion-based approach for multi-colored hair editing that effectively separates hair color and style control, leading to superior editing results while preserving facial attributes.

Contribution

The paper presents a multi-stage hairstyle blend and a warping module within latent diffusion models, enabling precise multi-color hair editing and attribute preservation, surpassing prior StyleGAN-based methods.

Findings

Outperforms existing methods in multi-color hairstyle editing

Effectively preserves facial attributes during editing

Demonstrates superior editing quality with textual and reference inputs

Abstract

Hair editing is a critical image synthesis task that aims to edit hair color and hairstyle using text descriptions or reference images, while preserving irrelevant attributes (e.g., identity, background, cloth). Many existing methods are based on StyleGAN to address this task. However, due to the limited spatial distribution of StyleGAN, it struggles with multiple hair color editing and facial preservation. Considering the advancements in diffusion models, we utilize Latent Diffusion Models (LDMs) for hairstyle editing. Our approach introduces Multi-stage Hairstyle Blend (MHB), effectively separating control of hair color and hairstyle in diffusion latent space. Additionally, we train a warping module to align the hair color with the target region. To further enhance multi-color hairstyle editing, we fine-tuned a CLIP model using a multi-color hairstyle dataset. Our method not only tackles the complexity of multi-color hairstyles but also addresses the challenge of preserving original colors during diffusion editing. Extensive experiments showcase the superiority of our method in editing multi-color hairstyles while preserving facial attributes given textual descriptions and reference images.