DreamMakeup: Face Makeup Customization using Latent Diffusion Models

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Park, Geon Yeong, Han, Inhwa, Yang, Serin, Hong, Yeobin, Jeong, Seongmin, Jeon, Heechan, Goh, Myeongjin, Yi, Sung Won, Nam, Jin, Ye, Jong Chul
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866917007266414592
author Park, Geon Yeong
Han, Inhwa
Yang, Serin
Hong, Yeobin
Jeong, Seongmin
Jeon, Heechan
Goh, Myeongjin
Yi, Sung Won
Nam, Jin
Ye, Jong Chul
author_facet Park, Geon Yeong
Han, Inhwa
Yang, Serin
Hong, Yeobin
Jeong, Seongmin
Jeon, Heechan
Goh, Myeongjin
Yi, Sung Won
Nam, Jin
Ye, Jong Chul
contents The exponential growth of the global makeup market has paralleled advancements in virtual makeup simulation technology. Despite the progress led by GANs, their application still encounters significant challenges, including training instability and limited customization capabilities. Addressing these challenges, we introduce DreamMakup - a novel training-free Diffusion model based Makeup Customization method, leveraging the inherent advantages of diffusion models for superior controllability and precise real-image editing. DreamMakeup employs early-stopped DDIM inversion to preserve the facial structure and identity while enabling extensive customization through various conditioning inputs such as reference images, specific RGB colors, and textual descriptions. Our model demonstrates notable improvements over existing GAN-based and recent diffusion-based frameworks - improved customization, color-matching capabilities, identity preservation and compatibility with textual descriptions or LLMs with affordable computational costs.
format Preprint
id arxiv_https___arxiv_org_abs_2510_10918
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle DreamMakeup: Face Makeup Customization using Latent Diffusion Models
Park, Geon Yeong
Han, Inhwa
Yang, Serin
Hong, Yeobin
Jeong, Seongmin
Jeon, Heechan
Goh, Myeongjin
Yi, Sung Won
Nam, Jin
Ye, Jong Chul
Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
The exponential growth of the global makeup market has paralleled advancements in virtual makeup simulation technology. Despite the progress led by GANs, their application still encounters significant challenges, including training instability and limited customization capabilities. Addressing these challenges, we introduce DreamMakup - a novel training-free Diffusion model based Makeup Customization method, leveraging the inherent advantages of diffusion models for superior controllability and precise real-image editing. DreamMakeup employs early-stopped DDIM inversion to preserve the facial structure and identity while enabling extensive customization through various conditioning inputs such as reference images, specific RGB colors, and textual descriptions. Our model demonstrates notable improvements over existing GAN-based and recent diffusion-based frameworks - improved customization, color-matching capabilities, identity preservation and compatibility with textual descriptions or LLMs with affordable computational costs.
title DreamMakeup: Face Makeup Customization using Latent Diffusion Models
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2510.10918