Makeup transfer dataset. This is an interesting but challenging task.
Makeup transfer dataset Using the novel network architecture and the newly collected datasets for training, we obtain an all-inclusive makeup transfer method that outperforms all previous meth-ods Sep 3, 2021 · About the Makeup Transfer dataset. We also introduce 4 new datasets (both real and synthesis) to train and evaluate CPM. This survey paper provides a comprehensive Nov 18, 2024 · The advancement of makeup transfer, editing, and image encoding has demonstrated their effectiveness and superior quality. It contains 3,000 high-quality face images with a higher resolution of 512*512, covering more recent makeup styles and more diverse face poses, backgrounds, expressions, races, illumination, etc. Existing lit-erature leverage the adversarial loss so that the gen-erated faces are of high quality and realistic as real ones, but are only able to produce fixed outputs. We also contribute BeautyFace, a makeup transfer dataset to supplement existing datasets. Our code uses Face++ Detection API for facial landmarks, and the downloaded dataset includes the facial landmarks of the dataset images. This is an interesting but challenging task. Dataset: Facial Makeup Transfer Dataset. Inspired by recent advances in disentangled repre-sentation, in this paper keup transfer in the real-world scenarios. In this study Abstract Current makeup transfer methods are limited to simple makeup styles, making them difficult to apply in real-world scenarios. Jun 13, 2024 · In this article, we address the challenging makeup transfer task, aiming to transfer makeup from a reference image to a source image while preserving facial geometry and background consistency. In this paper, we address the makeup transfer task, which aims to transfer the makeup from a reference image to a source image. Makeup transfer networks can translate the makeup style of a reference image to any other non-makeup one while preserving face identity, helping people find the most suitable makeup for them and get the beautified image. org/10. , Chen et al. Abstract Facial makeup transfer is a widely-used technology that aims to transfer the makeup style from a refer-ence face image to a non-makeup face. Besides, they cannot realize customizable transfer that allows a controllable shade of makeup Apr 1, 2025 · From a data perspective, existing makeup datasets lack diversity and cannot accommodate real-world makeup transfer. Gu et al. Source code for demo, web, etc. Such an instance-level transfer prob-lem is more challenging than conventional domain-level transfer tasks, especially when paired data is unavailable. Please feel free to send me pull requests (or issues) to add papers/ talks/ demo etc. A curated list of Makeup Transfer (and Hairstyles Transfer) and related resources. Futhermore, existing high-dimensional latent encoding methods mainly target global Abstract Existing makeup techniques often require designing multiple models to handle different inputs and align features across domains for different makeup tasks, e. This dataset is used for facial makeup transfer. They include extreme makeup styles, which do not exist in previous makeup datasets. In this paper, we construct the Peking Opera makeup dataset and propose a makeup transfer network for Opera faces called OperaGAN. Since then, many methods have been proposed to accomplish MT, most of which are based on Aug 5, 2022 · Instance-level makeup transfer is studied on the basis of domain-level makeup transfer, so it is necessary to consider two inter-domain datasets, the natural face dataset X α ⊂ R H × W × 3 and the reference face dataset Y β ⊂ R H × W × 3. Mar 1, 2025 · The texture details of headwear in style examples tend to be ignored. Due to the absence of paired data, current methods typically synthesize sub-optimal pseudo ground truths to guide the model training, resulting in low makeup fidelity. Oct 3, 2024 · Existing methods for makeup transfer mainly focus on the transfer of facial makeup, while overlooking the importance of hairstyles. Existing deep neural network-based methods have shown Abstract Makeup transfer is the task of applying on a source face the makeup style from a reference image. When dealing with complex portrait style transfer, simultaneous correct headwear and facial makeup transfer often cannot be guaranteed. The latter contains images with various postures and facial expressions. To enrich our training in advance, we also collected images manually, but due to the poor training performance, these data were discarded. Mar 12, 2024 · Current makeup transfer methods are limited to simple makeup styles, making them difficult to apply in real-world scenarios. Aug 6, 2025 · Built upon the diverse FFHQ dataset, our pipeline transfers real-world makeup styles from existing datasets onto 18K identities by introduc-ing an improved makeup transfer method that disentangles identity and makeup. Aug 1, 2023 · Makeup transfer (MT) aims to transfer the makeup style from a given reference makeup face image to a source image while preserving face identity and background information. The dataset consists of 4 images per subject: 1 before-makeup image and 3 after-makeup images. · Issue #11 · VinAIResearch/CPM · GitHub VinAIResearch / CPM Public Notifications Fork 64 Star 401 Mar 1, 2025 · The texture details of headwear in style examples tend to be ignored. To train and evaluate such a system, we also introduce new makeup datasets for real and synthetic extreme makeup. There is a need to pay attention to the process of MT: lipstick, eye shadow, foundation, blush etc. Since there is no paired dataset, we formulate a new loss function to guide the decomposition. Real-life makeups are diverse and wild, which cover not only color-changing but also patterns, such as stickers, blushes, and jewelries. ** CVPRW 2023 Paper Link: paper Apr 1, 2022 · In our experiment, we trained and evaluated RAMT-GAN on the Makeup Transfer dataset that contains unpaired makeup and non-makeup face images. Our model consists of two parallel branches: the facial makeup transfer branch We propose a local adversarial disentangling network (LADN) for facial makeup and de-makeup. In this paper, we introduce Stable-Makeup, a novel diffusion-based makeup transfer method capable of robustly transferring a wide range of real-world makeup, onto user-provided faces. Stable-Makeup is based on a pre-trained diffusion model and utilizes a Detail SCGAN Implementation of CVPR 2021 paper "Spatially-invariant Style-codes Controlled Makeup Transfer" A re-implementation of BeautyGAN: Instance-level Facial Makeup Transfer with Deep Generative Adversarial Network (ACM MM'18) - thaoshibe/BeautyGAN-PyTorch-reimplementation Mar 12, 2024 · Abstract Current makeup transfer methods are limited to simple makeup styles, making them difficult to apply in real-world scenarios. In addition, due to limited Facial makeup transfer aims to translate the makeup style from a given reference makeup face image to another non-makeup one while preserving face identity. However, BeautyGAN only translates the color distribution but not the makeup details. In recent years, MT has attracted the attention of many scholars, and it has a wide range of application prospects and research value. The key insights of this study are modeling component-specific correspondence for local makeup transfer, capturing long-range dependencies for global makeup transfer, and enabling eficient makeup transfer via a single-path structure. To address this issue, we propose a composite makeup transfer model based on generative adversarial networks, which cleverly achieves composite transfer of facial makeup and hairstyles. Existing methods have achieved much advancement in constrained scenarios, but it is still very challenging for them to transfer makeup between images with large pose and Mar 12, 2024 · Abstract Current makeup transfer methods are limited to simple makeup styles, making them difficult to apply in real-world scenarios. Extensive experiments show that BeautyGAN could generate visually pleasant makeup faces and accurate transferring results. Each identity is paired with 5 different makeup styles, resulting in a total of 90K high-quality bare–makeup image pairs. With advancements in generative models, information fusion—a process that integrates and interprets data from diverse sources—has gained new opportunities for improvement in tasks like makeup transfer. PSGAN [3], SCGAN [4], and TSEV-GAN [5 Mar 8, 2023 · Virtual makeup transfer has emerged as a rapidly growing area of research, fueled by advancements in deep learning , computer vision, and graphics. Additionally, different makeup styles generally have varying effects on the Oct 17, 2022 · Makeup transfer (MT) aims to transfer the makeup style from a given reference makeup face image to a source image while preserving face identity and background information. However, the above works are sensitive to the large pose and expression variations. The json representation of the dataset with its distributions based on DCAT. Dec 12, 2022 · The key insights of this study are modeling component-specific correspondence for local makeup transfer, capturing long-range dependencies for global makeup transfer, and enabling efficient makeup transfer via a single-path structure. BeautyGAN, a makeup transfer network The FRGC images undergo 3 virtual makeovers: (1) application of lipstick only, (2) application of eye makeup only, and (3) application of a full makeup suite consisting of lipstick, foundation, blush and eye makeup. You can prepare datasets following our paper and make a jsonl file (each line with 4 key-value Feb 1, 2025 · Specifically, BeautyGAN [1] is the pioneer of GAN-based makeup transfer models, which proposed the makeup loss and MT dataset. However, existing works overlooked the latter components and confined makeup transfer to color manipulation, fo-cusing only on light makeup styles. Existing methods have achieved much advancement in constrained scenarios, but it is still very challenging for them to transfer makeup between images with large pose and About transfer the makeup style of a reference face image to a non-makeup face histogram-matching makeup-transfer Readme Activity Our proposed framework, Stable-Makeup, is a novel diffusion-based method for makeup transfer that can robustly transfer a diverse range of real-world makeup styles, from light to extremely heavy makeup. FFHQ-Makeup is a large-scale paired synthetic facial makeup dataset designed to support research on virtual try-on, makeup transfer, and beauty-related vision tasks. CPM consists of an improved color transfer branch (based on BeautyGAN) and a novel pattern transfer branch. Thus, we collect a new Makeup-Heavy (MH) dataset that contains facial images with a wide range of heavy makeup s yles under various poses and ex-pressions. To be specific, we compare our approach against several previous makeup transfer works, including two classic image-to-image generation methods, DIA [23] and CycleGAN [5], and four advanced GAN-based Oct 17, 2022 · Abstract and Figures Makeup transfer (MT) aims to transfer the makeup style from a given reference makeup face image to a source image while preserving face identity and background information. Aug 6, 2025 · Built upon the diverse FFHQ dataset, our pipeline transfers real-world makeup styles from existing datasets onto 18K identities by introducing an improved makeup transfer method that disentangles identity and makeup. Jan 1, 2021 · Makeup transfer methods base on paired data sets requires the target image and the reference image pair before and after makeup, the three images should be obtained in the same posture and light. Then modify the skin texture and color difference between the reference face and the target face As one of the important ways to change the appearance of face image, makeup transfer has received more and more attention in recent years. CPM: Color-Pattern Makeup Transfer CPM is a holistic makeup transfer framework that outperforms previous state-of-the-art models on both light and extreme makeup styles. Existing techniques May 25, 2021 · In this paper, we address the makeup transfer and removal tasks simultaneously, which aim to transfer the makeup from a reference image to a source image and remove the makeup from the with-makeup image respectively. In Aug 5, 2025 · Built upon the diverse FFHQ dataset, our pipeline transfers real-world makeup styles from existing datasets onto 18K identities by introducing an improved makeup transfer method that disentangles identity and makeup. Existing methods have achieved much advancement in constrained scenarios, but it is still very challenging for them to transfer makeup between images with large pose and Also, to evaluate the effectiveness of the model, the authors collected an image dataset MT in the Wild (MT‐Wild) and a Makeup Transfer High‐Resolution dataset containing various poses and expressions. g. This repository contains the official implementation of the following paper: **BeautyREC:Robust, Efficient, and Component-Specific Makeup Transfer. Dec 13, 2022 · CPM: Color-Pattern Makeup Transfer CPM is a holistic makeup transfer framework that outperforms previous state-of-the-art models on both light and extreme makeup styles. Download makeup dataset We release a dataset containing unpaired images before- and after-makeup faces, together with the synthetic ground truth. Dec 1, 2024 · Furthermore, in response to the lack of high-resolution makeup transfer datasets, we have curated a collection of 9716 face images at a resolution of 1024 × 1024. It consists of an improved color transfer branch and a novel pattern transfer branch to learn all makeup properties, including color, shape, texture, and location. In addition, due to limited Oct 17, 2022 · 1 INTRODUCTION Facial makeup transfer (MT) aims to transfer makeup style from a given reference face image to another non-makeup face image while preserving the facial identity. However, existing makeup works primarily focus on low-dimensional features such as color distributions and patterns, limiting their versatillity across a wide range of makeup applications. In addition, we contacted the author of BeautyGAN [6] and were authorized to use their large Makeup Transfer dataset. 📢 New: We provide Qualitative Performane Comparisons online Built upon the diverse FFHQ dataset, our pipeline trans-fers real-world makeup styles from existing datasets onto 18K identities by introducing an improved makeup transfer method that disentangles identity and makeup. https://doi. Oct 15, 2018 · We also build up a new makeup dataset that consists of 3834 high-resolution face images. 57702/45ny7jzc. Central to our method are multiple and overlapping local adversarial discriminators in a content-style disentangling network for achieving local detail transfer between facial images, with the use of asymmetric loss functions for dramatic makeup styles with high-frequency details. , beauty filter, makeup transfer, and makeup removal, leading to increased complexity. Ultimately, our training data comprises 20k image pairs . Dec 15, 2024 · This paper studies the challenging task of makeup transfer, which aims to apply diverse makeup styles precisely and naturally to a given facial image. Existing methods have achieved promising progress in constrained scenarios, but transferring between images with large pose and expression differences is still challenging. Another limitation is the absence of text-guided makeup try-on, which is more user-friendly without needing reference images. Prior to the emergence of Sep 16, 2019 · In this paper, we address the makeup transfer task, which aims to transfer the makeup from a reference image to a source image. Firstly, calculate the changes in the image color and illumination before and after applying makeup. [2] proposed the complex style transfer. Since then, many methods have been proposed to accomplish MT, most of which are based on To facilitate on-demand makeup transfer, in this work, we pro-pose BeautyGlow that decompose the latent vectors of face images derived from the Glow model into makeup and non-makeup latent vectors. Nov 1, 2025 · Makeup transfer refers to applying the makeup style from a reference image to a source image while preserving the source’s facial features and background. (2024). Then, Gu et al. Oct 17, 2022 · 1 INTRODUCTION Facial makeup transfer (MT) aims to transfer makeup style from a given reference face image to another non-makeup face image while preserving the facial identity. Figure 1 shows the classic MT result. We collect 558 with-makeup images from the In-ternet, and the CPM is a holistic makeup transfer framework that outperforms previous state-of-the-art models on both light and extreme makeup styles. To address this data shortage, we propose an automatic data construction pipeline that employs a large language model and generative model to edit real human face images and create paired before-and-after makeup images. May 26, 2021 · In this paper, we address the makeup transfer and removal tasks simultaneously, which aim to transfer the makeup from a reference image to a source image and remove the makeup from the with-makeup image respectively. Makeup Transfer High Resolution (MT-HR) dataset is the other newly collected high- resolution makeup dataset with 2;000 with-makeup images and 1;000 non-makeup images in 512 512, which is used to We also introduce new makeup-transfer datasets, con-sisting of both synthetic and real images, and covering a wide range of makeup styles. transfer the makeup style of a reference face image to a non-makeup face Apr 1, 2022 · It is worth mentioning that PSGAN uses Makeup Transfer dataset and Makeup-Wild dataset to complete network training. ownajlxpoboizmqovqpiwqbvlbkgfkyyhvvgdeigolfhkeklxtkujxubjpgtbjuyujvduzalmke