Pith. sign in

REVIEW 4 major objections 5 minor 1 cited by

FFHQ-Makeup: Paired Synthetic Makeup Dataset with Facial Consistency Across Multiple Styles

T0 review · 4 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read The paper builds a synthetic dataset of 18K identities, each paired with five makeup styles, keeping identity and expression fixed across the bare–makeup pairs.

desk verdict A genuinely useful paired makeup dataset if it ships, with an overstated disentanglement story and thin evaluation. read the letter →

arxiv 2508.03241 v2 pith:A7NQKKBE submitted 2025-08-05 cs.CV

Xingchao Yang , Shiori Ueda , Yuantian Huang , Tomoya Akiyama , Takafumi Taketomi This is my paper
classification cs.CV
keywords FFHQ-Makeupsyntheticmakeupdatasettransfer3Dmorphablemodeldiffusionmodelsfacialconsistencypairedbare-makeupimagesfaceediting
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

FFHQ-Makeup is a large synthetic dataset of 90,000 paired bare–makeup face images: 18,000 identities from FFHQ, each paired with five different makeup styles at 512×512 resolution. The paper's goal is to break the trade-off that has made such pairs scarce: real before-and-after photos are costly to collect, warping-based synthesis distorts faces, and text-to-image synthesis drifts identity and expression. The authors' generation pipeline fits a 3D morphable model to a makeup image, reconstructs a bare version of the same face, subtracts it to isolate a makeup residual, and re-renders that residual onto many different target faces before training a diffusion makeup-transfer model. Their central claim is that this residual-based disentanglement, together with structure control from the reconstructed face and landmarks, produces bare–makeup pairs that change makeup while keeping identity and expression fixed, and the resulting dataset outperforms existing synthetic paired datasets on identity similarity and semantic consistency.

What carries the argument

The load-bearing object is the 3DMM-based makeup residual $R=I_S-\hat{I}_b$, where $\hat{I}_b$ is a bare face reconstructed by fitting a 3D morphable model (a statistical model of 3D face shape and texture) to the makeup image and blending it with the background. Subtracting the reconstruction strips away identity, expression, pose, and illumination cues, leaving appearance tied to makeup; sampling the residual's vertex-wise colors and re-rendering them onto many different reconstructed face geometries further enforces invariance to facial structure. The makeup-transfer network then reads these augmented residuals with a frozen CLIP image encoder and a channel-attention residual learner, while ControlNet conditions generation on the reconstructed bare face and detected landmarks. This separation of appearance features from structural guidance is what lets the pipeline generate 90K paired images while holding identity and expression fixed.

What would settle it

Take a makeup image, fit the 3DMM, render the bare reconstruction, and form the residual; feed the residual to the trained transfer model and generate makeup on a different identity. Then run ArcFace on the generated face against the original makeup-image identity: if this identity similarity is substantially above chance, the residual still encodes identity and the disentanglement claim is false. A simpler check is to render the residual colors onto a probe 3D face and ask an identity-recognition model whether it matches the source subject.

Watch

Extended reading notes

Core claim

On the paper's own terms, the core discovery is that faithful bare–makeup pairs can be synthesized at scale without any real paired supervision. For a makeup image $I_S$, the method fits a 3D morphable model to capture identity, expression, pose, illumination, and skin tone, renders a reconstructed bare face $\hat{I}_b$, blends it with the original background, and forms the makeup residual $R=I_S-\hat{I}_b$; because this residual lives on the reconstructed 3D surface, its color values can be sampled and re-rendered onto other faces. The network therefore never sees the residual attached to the source face's geometry, and the makeup style is decoupled from facial structure. Trained with this representation and guided by ControlNet on the reconstructed bare face and landmarks, the model transfers real makeup styles onto 18,000 FFHQ identities, five styles each, producing 90,000 pairs. The paper reports that these pairs achieve higher ArcFace identity similarity and DINO-I semantic consistency than the existing LADN-Syn and BeautyBank synthetic datasets.

Load-bearing premise

The whole pipeline rests on the assumption that fitting a single 3D morphable model to a makeup image reconstructs the bare face accurately enough that subtracting it leaves a residual containing the makeup but none of the person's identity or facial geometry.

Editorial extensions

If this is right

  • A public 90K-pair dataset is available for beauty-related tasks that previously had to train on unpaired or low-quality data.
  • Models for makeup-invariant face recognition and facial privacy protection can be trained and benchmarked on consistent bare–makeup pairs at scale.
  • Because each identity appears with five styles, the dataset supports identity-conditioned style transfer and controlled studies of how makeup affects recognition.
  • The pair-free generation pipeline can be re-run on new face datasets, limited mainly by 3DMM fitting and segmentation quality, so the dataset can grow beyond FFHQ.
  • Other synthetic makeup datasets can now be compared directly against FFHQ-Makeup on identity and semantic consistency metrics, giving the field a common evaluation set.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • A testable consequence of the disentanglement claim is that an identity-recognition model should not be able to match a residual-rendered face back to the original makeup image's identity; if it can, the residual is leaking identity.
  • The same residual sampling-and-re-rendering trick could in principle be applied to other appearance attributes that lack paired data, such as beard growth, skin texture, or aging, whenever a 3D prior can reconstruct the attribute-free version.
  • Because the style pool is limited to MT and LADN and the identity pool to FFHQ, the reported consistency gains should be re-tested when either pool is expanded, since the current numbers could partly reflect the curated, well-aligned source material.
  • Independent human raters, rather than a vision-language model, would strengthen the visual-preference comparison; if humans do not share the reported preference, the perceptual claim would need qualification.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 5 minor

Summary. The paper introduces FFHQ-Makeup, a synthetic paired bare-makeup dataset built from 18K FFHQ identities, each paired with five makeup styles for a total of 90K image pairs. The generation pipeline uses a 3DMM to reconstruct a bare face from a single makeup image, computes a makeup residual R = I_S - \hat I_b, augments the residual by sampling and re-rendering on diverse target geometries, and feeds the augmented residual through a Stable-Makeup/FreeUV-style diffusion pipeline with ControlNet for structural guidance. The authors claim that this disentangles identity from makeup, and they report quantitative and visual preference comparisons against LADN-Syn and BeautyBank, plus ablations. The paper's main contribution is the dataset itself, with the generation method supporting its construction.

Significance. If the dataset is released as promised, it would fill a real gap: existing paired synthetic makeup datasets are either small, not public, or suffer from identity/expression drift. The scale (18K identities, 5 styles per identity, 90K pairs), the multi-style per-identity design, and the public availability are concrete strengths. The paper also uses external metrics (ArcFace, DINO-I, SSIM) rather than relying only on its own outputs, and the ablation study is a useful check on the design choices. However, the central claim that the 3DMM residual is makeup-only is not directly measured, and the reported quantitative advantages are presented without variance or significance tests. The manual curation steps are also not specified, so the dataset-level comparison may partly reflect the filtering effort rather than the method. These issues are fixable and do not invalidate the resource itself, but they need to be addressed before the empirical claims can be accepted at face value.

major comments (4)
  1. [Section 3.1] The central disentanglement claim rests on the residual R = I_S - \hat I_b being a makeup-only signal, but the paper never measures how much identity, geometry, lighting, or fitting error remains in R. 3DMM fitting from a makeup image is ill-posed because makeup alters the observed albedo, so the optimizer may fold makeup into shape or shading, and source-specific structure such as nostrils, wrinkles, or illumination can leak into the residual. Since the sampling-and-re-rendering step transports R onto every target geometry, any non-makeup content becomes a source of cross-identity contamination in all 90K pairs. The ablations in Section 4.2 show that the residual representation helps, but they do not quantify residual purity. The paper's own limitation section (Section 5.1) admits that results are "partially affected by the accuracy of 3DMM fitting," yet no diagnostic quantifies this. I recommend adding a direct measurement: for example, train a classifier or linear probe on R to see whether source identity can be recovered, or report ArcFace similarity between residuals rendered on different geometries, or decompose the residual into makeup vs. non-makeup components on a small labeled set. Without such evidence, the superiority of FFHQ-Makeup in Table 3 could be attributed to the ControlNet structure prior rather than to a truly makeup-only residual.
  2. [Section 4.4, Table 3] The quantitative claims of superiority are based on single point estimates of ArcFace, DINO-I, and SSIM, with no variances, confidence intervals, or significance tests. This matters because the dataset-level comparison samples 90K images from each dataset, so even small differences could be statistically unreliable; the method-level rows in the same table are presumably on a smaller subset and may not be directly comparable. I recommend reporting bootstrap confidence intervals or per-identity/per-style error bars, and a paired significance test where appropriate (for example, when the same target identities are generated by different variants). Without this, "outperforms" in the abstract and conclusion is not yet supported.
  3. [Section 4.1, Table 2] The visual preference study uses 50 groups judged by GPT-4o, with no human validation, no inter-annotator agreement measure, and no statistical test. A vision-language model may have systematic preferences (for example, favoring smoother images or penalizing certain artifacts) that do not match human judgment, and 50 groups is a small sample. I recommend either adding a human preference study (even on a subset) or reporting agreement between GPT-4o and human raters, and providing confidence intervals for the preference percentages.
  4. [Section 3.3] The dataset construction pipeline relies on several manual filtering steps: removing extreme makeup styles, manually masking failed segmentation regions, filtering out FFHQ samples with inaccurate 3DMM fitting, and group-wise visual inspection of the final outputs. These steps are not described with explicit criteria, and the dataset-level evaluation in Section 4.1 and Table 3 is performed on the final filtered dataset. As a result, the comparison against LADN-Syn and BeautyBank may partly measure the curation effort rather than the generation method. I recommend documenting the filtering criteria, reporting the number of discarded samples at each stage, and, if feasible, running the dataset-level metrics on unfiltered outputs to separate the method's contribution from the manual selection effect.
minor comments (5)
  1. [Section 2.1] There is a typo in the text: "Stable-Stable-Makeup" should be "Stable-Makeup," and the sentence starting "Stable-Stable-Makeup [51] proposed..." is grammatically incomplete.
  2. [Section 1] The phrase "existing makeup datasets construction still fail short" should be revised to "existing makeup dataset construction still falls short," and the abstract contains "this is the first work that focuses specifically on constructing makeup dataset," which should be softened and made grammatical since prior works such as LADN-Syn and BeautyBank also construct synthetic datasets.
  3. [Figure 8] Figure 8 contains unlabeled numeric annotations (e.g., 74, 178, 59, 426, 68766) whose meaning is unclear; please explain them in the caption or text.
  4. [References] References [1] and [2] are the same paper (Alashkar et al., AAAI 2017, "Examples-rules guided deep neural network for makeup recommendation"); please merge them and use a single reference.
  5. [Section 3.1] The sentence "This self-supervised approach enables scalable and flexible for synthetic makeup dataset generation" is ungrammatical; consider "This self-supervised approach enables scalable and flexible synthetic makeup dataset generation."

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: core dataset-quality claims are supported by external metrics and internal ablations, not by definition or by load-bearing self-citation.

full rationale

The paper's central claim is that FFHQ-Makeup provides realistic, consistent bare-makeup pairs. The construction uses a 3DMM residual R = I_S - I_b and ControlNet guidance, but the evaluation of consistency (ArcFace, DINO-I, SSIM in Table 3, and the GPT-4o preference study in Table 2) relies on externally defined similarity metrics and comparisons against LADN-Syn and BeautyBank. No fitted parameter is renamed as a prediction: the method is trained on makeup images and then applied to FFHQ targets, and the reported metrics are measured on the resulting pairs. The disentanglement claim is supported by ablation variants (w/o makeup residual and w/o sample and re-rendering), which is an internal comparison rather than a definitional identity. Self-citations to FreeUV and BeautyBank are used as architectural baselines and prior datasets, not as evidence that the central consistency claim holds. The residual-purity assumption is an empirical limitation (admitted in Sec 5.1), not a circular step: if the 3DMM fitting leaks identity, the metrics would degrade, so the claim is falsifiable rather than forced. No circularity is therefore present.

Assumptions & free parameters 2 free parameters · 5 assumptions · 0 invented entities

The central claims rest on the accuracy of 3DMM fitting and on the validity of residual subtraction as a disentanglement mechanism. The evaluation additionally assumes that the chosen metrics and the GPT-4o judge capture the intended notions of realism and consistency. No new physical entities are introduced.

free parameters (2)
  • augmentation count per makeup image = 100
    Chosen in Sec. 3.1; the number of re-rendered residual samples per source image, affects training data diversity.
  • makeup styles per identity = 5
    Design choice for dataset composition in Sec. 3.3; affects scale and diversity but is not fitted.
assumptions (5)
  • domain assumption 3DMM fitting reconstructs a bare face with sufficient accuracy that the residual R = I_S - hat I_b captures makeup appearance without identity or structure.
    Invoked in Sec. 3.1 when defining the makeup residual; the central disentanglement claim rests on this subtraction.
  • domain assumption Sampling and re-rendering residuals across target 3DMM geometries removes source-face structural artifacts.
    Sec. 3.1 and ablation in Sec. 4.2; no formal proof, only qualitative comparison.
  • domain assumption ControlNet with hat I_b and landmarks as input preserves identity and expression during diffusion generation.
    Sec. 3.2; relies on the pretrained ControlNet/Stable Diffusion behavior.
  • domain assumption ArcFace, DINO-I, and SSIM are valid proxies for the claimed P_consistency.
    Sec. 4.4; no calibration against human perception, and SSIM is known not to capture semantic consistency well.
  • domain assumption The evaluation with GPT-4o preference reflects human judgment of makeup realism and facial consistency.
    Section 4.1; GPT-4o's preferences are used as a stand-in for human evaluation without validation.

how reviews work

0 comments
Cite this review

Pith. "Pith review of FFHQ-Makeup: Paired Synthetic Makeup Dataset with Facial Consistency Across Multiple Styles." pith.science (2026). https://pith.science/paper/A7NQKKBE

@misc{pith2026250803241,
  author       = {Pith},
  title        = {Pith review of: FFHQ-Makeup: Paired Synthetic Makeup Dataset with Facial Consistency Across Multiple Styles},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/A7NQKKBE}},
  note         = {Machine review of arXiv:2508.03241}
}
read the original abstract

Paired bare-makeup facial images are essential for a wide range of beauty-related tasks, such as virtual try-on, facial privacy protection, and facial aesthetics analysis. However, collecting high-quality paired makeup datasets remains a significant challenge. Real-world data acquisition is constrained by the difficulty of collecting large-scale paired images, while existing synthetic approaches often suffer from limited realism or inconsistencies between bare and makeup images. Current synthetic methods typically fall into two categories: warping-based transformations, which often distort facial geometry and compromise the precision of makeup; and text-to-image generation, which tends to alter facial identity and expression, undermining consistency. In this work, we present FFHQ-Makeup, a high-quality synthetic makeup dataset that pairs each identity with multiple makeup styles while preserving facial consistency in both identity and expression. 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. Each identity is paired with 5 different makeup styles, resulting in a total of 90K high-quality bare-makeup image pairs. To the best of our knowledge, this is the first work that focuses specifically on constructing a makeup dataset. We hope that FFHQ-Makeup fills the gap of lacking high-quality bare-makeup paired datasets and serves as a valuable resource for future research in beauty-related tasks.

Figures

Figures reproduced from arXiv: 2508.03241 by the authors.

Figure 1
Figure 1. Examples from our FFHQ-Makeup dataset, where each [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Examples of existing large-scale synthetic paired [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 3
Figure 3. Overview of the FFHQ-Makeup dataset generation [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: Our FFHQ-Makeup dataset inherits the diversity of FFHQ. As shown, it includes multiple bare-makeup pairs examples across [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]
Figure 5
Figure 5. Figure 5: Qualitative comparison of makeup datasets. Our FFHQ [PITH_FULL_IMAGE:figures/full_fig_p006_5.png]
Figure 6
Figure 6. Figure 6: Ablation study on feature extraction without using [PITH_FULL_IMAGE:figures/full_fig_p006_6.png]
Figure 8
Figure 8. Figure 8: Comparison of makeup transfer methods for dataset generation. Our method best preserves the target identity and expression [PITH_FULL_IMAGE:figures/full_fig_p007_8.png]

Discussion (0). Sign in to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. From Synthetic to Real: Toward Identity-Consistent Makeup Transfer with Synthetic and Real Data

    cs.CV 2026-05 unverdicted novelty 6.0 of 10

    The work creates identity-consistent synthetic makeup data via ConsistentBeauty and adapts models to real images using reinforcement learning in RealBeauty, achieving better identity preservation and real-world perfor...

Reference graph

Works this paper leans on

53 extracted references · 48 canonical work pages · cited by 1 Pith paper

  1. [1]

    Examples-rules guided deep neural network for makeup rec- ommendation

    Taleb Alashkar, Songyao Jiang, Shuyang Wang, and Yun Fu. Examples-rules guided deep neural network for makeup rec- ommendation. In AAAI 2017, pages 941–947, 2017. 3

    work page 2017

  2. [2]

    Examples-Rules guided deep neural network for makeup recommendation

    Taleb Alashkar, Songyao Jiang, Shuyang Wang, and Yun Fu. Examples-Rules guided deep neural network for makeup recommendation. In AAAI, 2017, pages 941–947, 2017. 1

    work page 2017

  3. [3]

    Geneavatar: Generic expression-aware vol- umetric head avatar editing from a single image

    Chong Bao, Yinda Zhang, Yuan Li, Xiyu Zhang, Bang- bang Yang, Hujun Bao, Marc Pollefeys, Guofeng Zhang, and Zhaopeng Cui. Geneavatar: Generic expression-aware vol- umetric head avatar editing from a single image. In CVPR, 2024, pages 8952–8963, 2024. 3

    work page 2024

  4. [4]

    A morphable model for the synthesis of 3D faces

    V olker Blanz and Thomas Vetter. A morphable model for the synthesis of 3D faces. In Proc. of SIGGRAPH 1999, pages 187–194, 1999. 3

    work page 1999

  5. [5]

    LEDITS++: limitless image editing us- ing text-to-image models

    Manuel Brack, Felix Friedrich, Katharina Kornmeier, Linoy Tsaban, Patrick Schramowski, Kristian Kersting, and Apolin´ario Passos. LEDITS++: limitless image editing us- ing text-to-image models. InCVPR, 2024, pages 8861–8870,

    work page 2024

  6. [6]

    Emerg- ing properties in self-supervised vision transformers

    Mathilde Caron, Hugo Touvron, Ishan Misra, Herv ´e J´egou, Julien Mairal, Piotr Bojanowski, and Armand Joulin. Emerg- ing properties in self-supervised vision transformers. In ICCV 2021, pages 9630–9640, 2021. 7

    work page 2021

  7. [7]

    Auto- matic facial makeup detection with application in face recog- nition

    Cunjian Chen, Antitza Dantcheva, and Arun Ross. Auto- matic facial makeup detection with application in face recog- nition. In ICB, 2013, pages 1–8. IEEE, 2013. 1, 2, 3

    work page 2013

  8. [8]

    Spoofing faces using makeup: An investigative study

    Cunjian Chen, Antitza Dantcheva, Thomas Swearingen, and Arun Ross. Spoofing faces using makeup: An investigative study. In ISBA, 2017, pages 1–8. IEEE, 2017. 1, 2, 3

    work page 2017

Show all 53 references
  1. [9]

    BeautyGlow: On- demand makeup transfer framework with reversible genera- tive network

    Hung-Jen Chen, Ka-Ming Hui, Szu-Yu Wang, Li-Wu Tsao, Hong-Han Shuai, and Wen-Huang Cheng. BeautyGlow: On- demand makeup transfer framework with reversible genera- tive network. In CVPR 2019, pages 10042–10050, 2019. 3

    2019

  2. [10]

    Can facial cosmetics affect the matching accuracy of face recognition systems? In BTAS, 2012, pages 391–398, 2012

    Antitza Dantcheva, Cunjian Chen, and Arun Ross. Can facial cosmetics affect the matching accuracy of face recognition systems? In BTAS, 2012, pages 391–398, 2012. 1, 2, 3, 4

    2012

  3. [11]

    Spatially-invariant style-codes controlled makeup transfer

    Han Deng, Chu Han, Hongmin Cai, Guoqiang Han, and Shengfeng He. Spatially-invariant style-codes controlled makeup transfer. In CVPR 2021, pages 6549–6557, 2021. 3, 6

    2021

  4. [12]

    ArcFace: Additive angular margin loss for deep face recognition

    Jiankang Deng, Jia Guo, Niannan Xue, and Stefanos Zafeiriou. ArcFace: Additive angular margin loss for deep face recognition. In CVPR 2019, pages 4690–4699, 2019. 7

    2019

  5. [13]

    DiffusionRig: Learning personalized priors for facial appearance editing

    Zheng Ding, Xuaner Zhang, Zhihao Xia, Lars Jebe, Zhuowen Tu, and Xiuming Zhang. DiffusionRig: Learning personalized priors for facial appearance editing. In CVPR 2023, pages 12736–12746, 2023. 3

    2023

  6. [14]

    Bernhard Egger, William A. P. Smith, Ayush Tewari, Ste- fanie Wuhrer, Michael Zollh ¨ofer, Thabo Beeler, Florian Bernard, Timo Bolkart, Adam Kortylewski, Sami Romdhani, Christian Theobalt, V olker Blanz, and Thomas Vetter. 3D morphable face models - past, present, and future. ...

    2020

  7. [15]

    LADN: Local adversarial disentangling network for facial makeup and de-makeup

    Qiao Gu, Guanzhi Wang, Mang Tik Chiu, Yu-Wing Tai, and Chi-Keung Tang. LADN: Local adversarial disentangling network for facial makeup and de-makeup. In ICCV 2019, pages 10480–10489, 2019. 1, 2, 3, 4, 5, 8

    2019

  8. [16]

    Denoising diffu- sion probabilistic models

    Jonathan Ho, Ajay Jain, and Pieter Abbeel. Denoising diffu- sion probabilistic models. In NeurIPS 2020, 2020. 3

    2020

  9. [17]

    Makeup- robust face verification

    Junlin Hu, Yongxin Ge, Jiwen Lu, and Xin Feng. Makeup- robust face verification. In ICASSP , 2013, pages 2342–2346. IEEE, 2013. 1, 2, 3

    2013

  10. [18]

    Protecting facial pri- vacy: Generating adversarial identity masks via style-robust makeup transfer

    Shengshan Hu, Xiaogeng Liu, Yechao Zhang, Minghui Li, Leo Yu Zhang, Hai Jin, and Libing Wu. Protecting facial pri- vacy: Generating adversarial identity masks via style-robust makeup transfer. In CVPR, 2022, pages 14994–15003, 2022. 1, 3

    2022

  11. [19]

    Goucher, Adam Perel- man, and et al

    Aaron Hurst, Adam Lerer, Adam P. Goucher, Adam Perel- man, and et al. Gpt-4o system card. CoRR, abs/2410.21276,

    arXiv

  12. [20]

    Controlface: Harnessing facial parametric control for face rigging

    Wooseok Jang, Youngjun Hong, Geonho Cha, and Seungry- ong Kim. Controlface: Harnessing facial parametric control for face rigging. In CVPR, 2025, pages 5614–5624, 2025. 3

    2025

  13. [21]

    PSGAN: Pose and expression robust spatial-aware GAN for customizable makeup transfer

    Wentao Jiang, Si Liu, Chen Gao, Jie Cao, Ran He, Jiashi Feng, and Shuicheng Yan. PSGAN: Pose and expression robust spatial-aware GAN for customizable makeup transfer. In CVPR 2020, pages 5193–5201, 2020. 1, 2, 3, 4, 6

    2020

  14. [22]

    To- ward Tiny and High-Quality Facial Makeup with Data Am- plify Learning

    Qiaoqiao Jin, Xuanhong Chen, Meiguang Jin, Ying Chen, Rui Shi, Yucheng Zheng, Yupeng Zhu, and Bingbing Ni. To- ward Tiny and High-Quality Facial Makeup with Data Am- plify Learning. In ECCV 2024, pages 340–356, 2024. 3

    2024

  15. [23]

    A style-based generator architecture for generative adversarial networks

    Tero Karras, Samuli Laine, and Timo Aila. A style-based generator architecture for generative adversarial networks. In CVPR 2019, pages 4401–4410, 2019. 2, 4, 8

    2019

  16. [24]

    Black, Hao Li, and Javier Romero

    Tianye Li, Timo Bolkart, Michael J. Black, Hao Li, and Javier Romero. Learning a model of facial shape and ex- pression from 4D scans. ACM Transactions on Graphics, 36 (6):194:1–194:17, 2017. 3

    2017

  17. [25]

    BeautyGAN: Instance-level fa- cial makeup transfer with deep generative adversarial net- work

    Tingting Li, Ruihe Qian, Chao Dong, Si Liu, Qiong Yan, Wenwu Zhu, and Liang Lin. BeautyGAN: Instance-level fa- cial makeup transfer with deep generative adversarial net- work. In Proceedings of International Conference on Multi- media, pages 645–653, 2018. 1, 2, 3, 4, 8

    2018

  18. [26]

    Anti-Makeup: Learning A bi-level adversarial network for makeup-invariant face verification

    Yi Li, Lingxiao Song, Xiang Wu, Ran He, and Tieniu Tan. Anti-Makeup: Learning A bi-level adversarial network for makeup-invariant face verification. In AAAI, 2018 , pages 7057–7064, 2018. 1, 3

    2018

  19. [27]

    Photomaker: Customizing re- alistic human photos via stacked ID embedding

    Zhen Li, Mingdeng Cao, Xintao Wang, Zhongang Qi, Ming- Ming Cheng, and Ying Shan. Photomaker: Customizing re- alistic human photos via stacked ID embedding. In CVPR, 2024, pages 8640–8650, 2024. 3

    2024

  20. [28]

    Face beautification: Beyond makeup transfer

    Xudong Liu, Ruizhe Wang, Hao Peng, Minglei Yin, Chih- Fan Chen, and Xin Li. Face beautification: Beyond makeup transfer. Frontiers Comput. Sci., 4, 2022. 3

    2022

  21. [29]

    BeautyBank: Encoding facial makeup in latent space

    Qianwen Lu, Xingchao Yang, and Takafumi Taketomi. BeautyBank: Encoding facial makeup in latent space. In WACV , 2025, pages 4183–4193. IEEE, 2025. 2, 3, 4, 5

    2025

  22. [30]

    SOGAN: 3D-aware shadow and occlusion robust GAN for makeup transfer

    Yueming Lyu, Jing Dong, Bo Peng, Wei Wang, and Tieniu Tan. SOGAN: 3D-aware shadow and occlusion robust GAN for makeup transfer. In Proceedings of International Confer- ence on Multimedia, pages 3601–3609, 2021

    2021

  23. [31]

    Lipstick ain’t enough: Beyond color matching for in-the-wild makeup transfer

    Thao Nguyen, Anh Tuan Tran, and Minh Hoai. Lipstick ain’t enough: Beyond color matching for in-the-wild makeup transfer. In CVPR 2021, pages 13305–13314, 2021. 1, 2, 3, 4

    2021

  24. [32]

    DiFaReli: Diffusion face relighting

    Puntawat Ponglertnapakorn, Nontawat Tritrong, and Supa- sorn Suwajanakorn. DiFaReli: Diffusion face relighting. In ICCV 2023, pages 22589–22600, 2023. 3

    2023

  25. [33]

    Learning transferable visual models from natural language supervision

    Alec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh, Gabriel Goh, Sandhini Agarwal, Girish Sastry, Amanda Askell, Pamela Mishkin, Jack Clark, Gretchen Krueger, and Ilya Sutskever. Learning transferable visual models from natural language supervision. In ICML 2021, pages ...

    2021

  26. [34]

    High-Resolution image synthesis with latent diffusion models

    Robin Rombach, Andreas Blattmann, Dominik Lorenz, Patrick Esser, and Bj ¨orn Ommer. High-Resolution image synthesis with latent diffusion models. In CVPR 2022, pages 10674–10685, 2022. 3, 4

    2022

  27. [35]

    Data augmentation-assisted makeup-invariant face recognition

    Muhammad Sajid, Nouman Ali, Saadat Hanif Dar, Naeem Iqbal Ratyal, Asif Raza Butt, Bushra Zafar, Tamoor Shafique, Mirza Jabbar Aziz Baig, Imran Riaz, and Shah- baz Baig. Data augmentation-assisted makeup-invariant face recognition. Math. Prob. Eng., 2018(1):2850632, 2018. 1, 3

    2018

  28. [36]

    SSAT: A symmetric semantic-aware transformer network for makeup transfer and removal

    Zhaoyang Sun, Yaxiong Chen, and Shengwu Xiong. SSAT: A symmetric semantic-aware transformer network for makeup transfer and removal. In AAAI, 2022, pages 2325– 2334, 2022. 1, 3

    2022

  29. [37]

    SHMT: Self- supervised hierarchical makeup transfer via latent diffusion models

    Zhaoyang Sun, Shengwu Xiong, Yaxiong Chen, Fei Du, Weihua Chen, Fan Wang, and Yi Rong. SHMT: Self- supervised hierarchical makeup transfer via latent diffusion models. In NIPS 2024, pages 16016–16042, 2024. 6

    2024

  30. [38]

    Content-Style Decoupling for Unsupervised Makeup Transfer without Generating Pseudo Ground Truth

    Zhaoyang Sun, Shengwu Xiong, Yaxiong Chen, and Yi Rong. Content-Style Decoupling for Unsupervised Makeup Transfer without Generating Pseudo Ground Truth. InCVPR 2024, pages 7601–7610, 2024. 3, 6

    2024

  31. [39]

    LEDITS: Real image editing with ddpm inversion and semantic guidance

    Linoy Tsaban and Apolin ´ario Passos. LEDITS: Real image editing with ddpm inversion and semantic guidance. CoRR, abs/2307.00522, 2023. 2, 3

    2023 arXiv

  32. [40]

    InstantID: Zero-shot identity-preserving generation in seconds

    Qixun Wang, Xu Bai, Haofan Wang, Zekui Qin, Anthony Chen, Huaxia Li, Xu Tang, and Yao Hu. InstantID: Zero-shot identity-preserving generation in seconds. CoRR, abs/2401.07519, 2024. 3

    2024 arXiv

  33. [41]

    Fm2u-net: Face morphological multi-branch network for makeup-invariant face verification

    Wenxuan Wang, Yanwei Fu, Xuelin Qian, Yu-Gang Jiang, Qi Tian, and Xiangyang Xue. Fm2u-net: Face morphological multi-branch network for makeup-invariant face verification. In CVPR, 2020, pages 5729–5739, 2020. 3

    2020

  34. [42]

    Beauty3dfacenet: Deep geometry and texture fusion for 3d facial attractiveness prediction

    Qinjie Xiao, You Wu, Dinghong Wang, Yong-Liang Yang, and Xiaogang Jin. Beauty3dfacenet: Deep geometry and texture fusion for 3d facial attractiveness prediction. Com- put. Graph., 98:11–18, 2021. 1, 3

    2021

  35. [43]

    BeautyREC: Robust, ef- ficient, and component-specific makeup transfer

    Qixin Yan, Chunle Guo, Jixin Zhao, Yuekun Dai, Chen Change Loy, and Chongyi Li. BeautyREC: Robust, ef- ficient, and component-specific makeup transfer. InCVPRW, 2023, pages 1102–1110, 2023. 1, 2, 3, 6

    2023

  36. [44]

    EleGANt: Exquisite and locally editable GAN for makeup transfer

    Chenyu Yang, Wanrong He, Yingqing Xu, and Yang Gao. EleGANt: Exquisite and locally editable GAN for makeup transfer. In ECCV 2022, 2022. 3, 6

    2022

  37. [45]

    BareSkinNet: De- makeup and De-lighting via 3D Face Reconstruction

    Xingchao Yang and Takafumi Taketomi. BareSkinNet: De- makeup and De-lighting via 3D Face Reconstruction. Com- puter Graphics Forum, 41(7):623–634, 2022. 3

    2022

  38. [46]

    Makeup extraction of 3D representation via illumination-aware image decomposition

    Xingchao Yang, Takafumi Taketomi, and Yoshihiro Kanamori. Makeup extraction of 3D representation via illumination-aware image decomposition. Computer Graph- ics Forum, 42(2):293–307, 2023. 4

    2023

  39. [47]

    Makeup prior models for 3D facial makeup estimation and applications

    Xingchao Yang, Takafumi Taketomi, Yuki Endo, and Yoshi- hiro Kanamori. Makeup prior models for 3D facial makeup estimation and applications. In CVPR 2024 , pages 2165– 2175, 2024. 3

    2024

  40. [48]

    Freeuv: Ground-Truth-Free realistic facial UV texture recovery via cross-assembly inference strategy

    Xingchao Yang, Takafumi Taketomi, Yuki Endo, and Yoshi- hiro Kanamori. Freeuv: Ground-Truth-Free realistic facial UV texture recovery via cross-assembly inference strategy. In CVPR, 2025, pages 326–337, 2025. 3, 4

    2025

  41. [49]

    IP- Adapter: Text compatible image prompt adapter for text-to- image diffusion models

    Hu Ye, Jun Zhang, Sibo Liu, Xiao Han, and Wei Yang. IP- Adapter: Text compatible image prompt adapter for text-to- image diffusion models. CoRR, abs/2308.06721, 2023. 3

    2023 arXiv

  42. [50]

    Adding conditional control to text-to-image diffusion models

    Lvmin Zhang, Anyi Rao, and Maneesh Agrawala. Adding conditional control to text-to-image diffusion models. In ICCV 2023, pages 3813–3824, 2023. 3, 5

    2023

  43. [51]

    Stable- Makeup: When real-world makeup transfer meets diffusion model

    Yuxuan Zhang, Lifu Wei, Qing Zhang, Yiren Song, Jiaming Liu, Huaxia Li, Xu Tang, Yao Hu, and Haibo Zhao. Stable- Makeup: When real-world makeup transfer meets diffusion model. CoRR, abs/2403.07764, 2024. 1, 2, 3, 4, 5, 6

    2024 arXiv

  44. [52]

    DiffSwap: High-Fidelity and Con- trollable Face Swapping via 3D-Aware Masked Diffusion

    Wenliang Zhao, Yongming Rao, Weikang Shi, Zuyan Liu, Jie Zhou, and Jiwen Lu. DiffSwap: High-Fidelity and Con- trollable Face Swapping via 3D-Aware Masked Diffusion. In CVPR 2023, pages 8568–8577, 2023. 3

    2023

  45. [53]

    State of the art on monocular 3D face reconstruction, tracking, and applica- tions

    Michael Zollh ¨ofer, Justus Thies, Pablo Garrido, Derek Bradley, Thabo Beeler, Patrick P ´erez, Marc Stamminger, Matthias Nießner, and Christian Theobalt. State of the art on monocular 3D face reconstruction, tracking, and applica- tions. Computer Graphics Forum, 37(2):523–550...

    2018

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.