pith. sign in

arxiv: 2606.20094 · v1 · pith:3K6LV3W7new · submitted 2026-06-18 · 💻 cs.CV · cs.AI· cs.GR· cs.LG· cs.MM

MakeupMirror: Improving Facial Attribute Preservation in Diffusion Models for Makeup Transfer

Nefeli Andreou , Angel Mart\'inez-Gonz\'alez , Sabine Sternig , Matthieu Guillaumin , Epameinondas Antonakos , Michael Opitz This is my paper

Pith reviewed 2026-06-26 17:58 UTC · model grok-4.3

classification 💻 cs.CV cs.AIcs.GRcs.LGcs.MM
keywords makeup transferdiffusion modelsfacial identity preservationskin tone preservationControlNetvirtual try-onimage editing
0
0 comments X

The pith

MakeupMirror improves facial identity and skin tone preservation in diffusion models for makeup transfer.

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

The paper introduces MakeupMirror as a diffusion-based makeup transfer method that targets the identity and skin color drift problems in earlier models like Stable-Makeup. It adds four technical pieces: facial geometry guidance through ControlNets, separate controls for different face regions, explicit skin tone modulation during transfer, and a faster Langevin sampler. Tests on three datasets, including a new more diverse one, report a 60 percent gain in face recognition similarity, a 50 percent drop in skin tone difference, 0.7-second inference, and 94 percent expert approval on identity criteria. If these gains hold, virtual makeup try-on could move from research demos to reliable online shopping tools.

Core claim

MakeupMirror integrates facial geometry conditioning with ControlNets, region-specific makeup transfer control, skin tone-based makeup transfer modulation, and a Levenberg-Marquardt Langevin sampler to maintain facial fidelity and prevent skin tone alteration while transferring makeup in diffusion models.

What carries the argument

The four-way conditioning stack of ControlNet facial geometry, region-specific makeup controls, skin tone modulation, and optimized Langevin sampling that together steer the diffusion process toward identity-preserving outputs.

Load-bearing premise

The evaluation metrics and newly collected MakeupSelfies dataset provide an unbiased and representative measure of facial attribute preservation that generalizes beyond the tested scenarios.

What would settle it

A follow-up test on a held-out set of faces with different demographics or lighting that shows no measurable gain in facial recognition similarity or skin tone difference relative to Stable-Makeup.

Figures

Figures reproduced from arXiv: 2606.20094 by Angel Mart\'inez-Gonz\'alez, Epameinondas Antonakos, Matthieu Guillaumin, Michael Opitz, Nefeli Andreou, Sabine Sternig.

Figure 1
Figure 1. Figure 1: Makeup transfer, qualitative comparison. Makeup transfer is the task of applying a reference makeup (left image) to a source input image (second column). While both methods achieve photorealism and faithful makeup transfer, MakeupMirror (third column) better preserves facial attributes and skin tone of the source image compared to Stable-Makeup [45] (right image). which customers could faithfully visualize… view at source ↗
Figure 2
Figure 2. Figure 2: Illustration of the MakeupMirror architecture. Building upon Stable-Makeup, our approach enhances facial feature preserva￾tion by adding controller networks for Depth-Anything [42] and Canny edge [2] maps. We also adapt noise to facial segmentation regions and modulate the transfer strength based on the estimated skin tone difference between the source image and the reference look. dressed in these methods… view at source ↗
Figure 3
Figure 3. Figure 3: Sample images (source images and reference makeup image) from the herein collected MakeupSelfies dataset. [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Skin tone distributions of our datasets (top, middle) com [PITH_FULL_IMAGE:figures/full_fig_p007_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Qualitative results of each of our improvements compared to StableMakeup (column 5). StableMakeup suffers from changing [PITH_FULL_IMAGE:figures/full_fig_p008_5.png] view at source ↗
read the original abstract

Makeup transfer models enable fun augmented reality (AR) experiences as well as virtual try-on (VTO) for online makeup shopping. While recent state-of-the-art diffusion based solutions such as Stable-Makeup dramatically improve the accuracy and realism of makeup transfer, they still face limitations in identity and skin color preservation, making production-level VTO for makeup shopping unrealistic. In this work, we propose MakeupMirror, a diffusion-based approach to makeup transfer that makes significant progress towards preserving facial features and skin tone. We introduce several technical innovations over Stable-Makeup: (1) integration of facial geometry conditioning with ControlNets to maintain facial fidelity; (2) region-specific makeup transfer control to enable precise makeup application across facial regions such as skin, eyes and lips; (3) skin tone-based makeup transfer modulation that prevent skin tone alteration in cross-subject transfer scenarios; and (4) integration of a Levenberg-Marquardt Langevin sampler to speed up inference while maintaining generation quality. Our experiments on CPM-Real, Makeup Wild, and (herein newly collected, more diverse) MakeupSelfies datasets show that MakeupMirror improves relative facial recognition similarity by +60%, reduces relative skin tone difference by -50% over Stable-Makeup, with a latency of 0.7s, while achieving expert acceptance rate of 94% across core facial identity preservation criteria.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 0 minor

Summary. The manuscript introduces MakeupMirror, a diffusion-based makeup transfer model that builds on Stable-Makeup with four technical additions: facial geometry conditioning via ControlNets, region-specific makeup transfer control, skin-tone-based modulation to avoid cross-subject color shifts, and a Levenberg-Marquardt Langevin sampler for faster inference. On CPM-Real, Makeup Wild, and a newly collected MakeupSelfies dataset, the method is reported to deliver +60% relative facial-recognition similarity, -50% relative skin-tone difference, 0.7 s latency, and 94% expert acceptance on identity-preservation criteria.

Significance. If the reported gains are reproducible under controlled conditions, the work would constitute a practical step toward production-grade virtual try-on by directly targeting the identity and skin-tone failure modes that currently limit diffusion-based AR makeup applications.

major comments (1)
  1. The abstract (and therefore the central empirical claim) presents performance deltas without any description of experimental protocols, statistical tests, number of runs, or controls for confounding factors such as dataset composition or prompt engineering; this absence prevents assessment of whether the +60% / -50% figures are load-bearing or artifactual.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for highlighting the need for greater transparency in how the central empirical claims are presented. We address the comment below and commit to revisions that improve clarity without changing the reported results.

read point-by-point responses

  1. Referee: The abstract (and therefore the central empirical claim) presents performance deltas without any description of experimental protocols, statistical tests, number of runs, or controls for confounding factors such as dataset composition or prompt engineering; this absence prevents assessment of whether the +60% / -50% figures are load-bearing or artifactual.

    Authors: We agree that the abstract, as currently written, does not describe the experimental protocols, statistical tests, number of runs, or explicit controls for confounding factors. The manuscript body (Section 4) details the three datasets, metrics, and comparison to Stable-Makeup, but the abstract itself provides no such context. We will revise the abstract to include a concise statement of the evaluation protocol (datasets, fixed reference/prompt setup, and identity/skin-tone metrics) and will add a pointer to the Experiments section. We will also ensure the Experiments section explicitly states the number of runs, any statistical testing performed, and controls applied for dataset composition and prompt engineering. These changes will be made in the next revision. revision: yes

Circularity Check

0 steps flagged

No significant circularity detected

full rationale

The abstract and available context present empirical comparisons of MakeupMirror against Stable-Makeup on CPM-Real, Makeup Wild, and MakeupSelfies datasets, reporting quantitative improvements in facial recognition similarity, skin tone difference, latency, and expert acceptance. No derivation chain, equations, fitted parameters renamed as predictions, or self-citation load-bearing steps are described. The central claims rest on experimental results rather than any self-referential construction or ansatz smuggled via citation, making the work self-contained against external benchmarks with no reduction of outputs to inputs by definition.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract provides no details on any free parameters, axioms, or invented entities used in the method.

pith-pipeline@v0.9.1-grok · 5811 in / 1028 out tokens · 31959 ms · 2026-06-26T17:58:07.707627+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

48 extracted references · 13 canonical work pages · 5 internal anchors

  1. [1]

    Stable Video Diffusion: Scaling Latent Video Diffusion Models to Large Datasets

    A. Blattmann, Tim Dockhorn, Sumith Kulal, Daniel Mendelevitch, Maciej Kilian, and Dominik Lorenz. Sta- ble video diffusion: Scaling latent video diffusion models to large datasets.ArXiv, abs/2311.15127, 2023. 2

    work page internal anchor Pith review Pith/arXiv arXiv 2023

  2. [2]

    A computational approach to edge detection

    John Canny. A computational approach to edge detection. IEEE Transactions on Pattern Analysis and Machine Intelli- gence, PAMI-8(6):679–698, 1986. 2, 3, 4

    1986

  3. [3]

    Diffusion models beat gans on image synthesis.Advances in Neural Informa- tion Processing Systems, 34, 2021

    Prafulla Dhariwal and Alexander Nichol. Diffusion models beat gans on image synthesis.Advances in Neural Informa- tion Processing Systems, 34, 2021. 2

    2021

  4. [4]

    Efros, and Aleksander Holynski

    Dave Epstein, Allan Jabri, Ben Poole, Alexei A. Efros, and Aleksander Holynski. Diffusion self-guidance for control- lable image generation. InProceedings of the 37th Inter- national Conference on Neural Information Processing Sys- tems, Red Hook, NY , USA, 2023. Curran Associates Inc. 2

    2023

  5. [5]

    Ladn: Local adversarial disentangling net- work for facial makeup and de-makeup

    Qiao Gu, Guanzhi Wang, Mang Tik Chiu, Yu-Wing Tai, and Chi-Keung Tang. Ladn: Local adversarial disentangling net- work for facial makeup and de-makeup. InProceedings of the IEEE/CVF International conference on computer vision, pages 10481–10490, 2019. 2

    2019

  6. [6]

    Digital face makeup by exam- ple.2009 IEEE Conference on Computer Vision and Pattern Recognition, pages 73–79, 2009

    Dong Guo and Terence Sim. Digital face makeup by exam- ple.2009 IEEE Conference on Computer Vision and Pattern Recognition, pages 73–79, 2009. 1, 2

    2009

  7. [7]

    Classifier-free diffusion guidance

    Jonathan Ho and Tim Salimans. Classifier-free diffusion guidance. InNeurIPS 2021 Workshop on Deep Generative Models and Downstream Applications, 2021. 2, 4

    2021

  8. [8]

    Denoising diffu- sion probabilistic models.Advances in Neural Information Processing Systems, 33:6840–6851, 2020

    Jonathan Ho, Ajay Jain, and Pieter Abbeel. Denoising diffu- sion probabilistic models.Advances in Neural Information Processing Systems, 33:6840–6851, 2020. 2

    2020

  9. [9]

    Psgan: Pose and expression ro- bust 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 ro- bust spatial-aware gan for customizable makeup transfer. In Proceedings of the IEEE/CVF conference on computer vi- sion and pattern recognition, pages 5194–5202, 2020. 1, 2, 5, 7

    2020

  10. [10]

    Physically-based simulation of cosmetics via intrinsic image decomposition with facial priors.IEEE Trans

    Chen Li, Kun Zhou, Hsiang-Tao Wu, and Stephen Lin. Physically-based simulation of cosmetics via intrinsic image decomposition with facial priors.IEEE Trans. Pattern Anal. Mach. Intell., 41(6):1455–1469, 2019. 1, 2

    2019

  11. [11]

    Layerdiffusion: Layered controlled image editing with dif- fusion models

    Pengzhi Li, Qinxuan Huang, Yikang Ding, and Zhiheng Li. Layerdiffusion: Layered controlled image editing with dif- fusion models. InSIGGRAPH Asia 2023 Technical Commu- nications, New York, NY , USA, 2023. Association for Com- puting Machinery. 2

    2023

  12. [12]

    Beautygan: Instance-level facial makeup transfer with deep generative adversarial network

    Tingting Li, Ruihe Qian, Chao Dong, Si Liu, Qiong Yan, Wenwu Zhu, and Liang Lin. Beautygan: Instance-level facial makeup transfer with deep generative adversarial network. InProceedings of the 26th ACM International Conference on Multimedia, page 645–653, New York, NY , USA, 2018. Association for Computing Machinery. 2, 5, 7

    2018

  13. [13]

    wow! you are so beautiful today!

    Luoqi Liu, Junliang Xing, Si Liu, Hui Xu, Xi Zhou, and Shuicheng Yan. “wow! you are so beautiful today!”.ACM Trans. Multimedia Comput. Commun. Appl., 11(1s), 2014. 1, 2

    2014

  14. [14]

    Makeupdiffuse: a double image-controlled diffusion model for exquisite makeup transfer.The Visual Computer, pages 1–17, 2024

    Xiongbo Lu, Feng Liu, Yi Rong, Yaxiong Chen, and Shengwu Xiong. Makeupdiffuse: a double image-controlled diffusion model for exquisite makeup transfer.The Visual Computer, pages 1–17, 2024. 2, 3

    2024

  15. [15]

    Mediapipe: A framework for perceiving and processing reality

    Camillo Lugaresi, Jiuqiang Tang, Hadon Nash, Chris Mc- Clanahan, Esha Uboweja, Michael Hays, Fan Zhang, Chuo- Ling Chang, Ming Yong, Juhyun Lee, Wan-Teh Chang, Wei Hua, Manfred Georg, and Matthias Grundmann. Mediapipe: A framework for perceiving and processing reality. InThird Workshop on Computer Vision for AR/VR at IEEE Computer Vision and Pattern Recog...

    2019

  16. [16]

    Repaint: Inpaint- ing using denoising diffusion probabilistic models.arXiv preprint arXiv:2201.09865, 2022

    Andreas Lugmayr, Martin Danelljan, Andres Romero, Fisher Yu, Radu Timofte, and Luc Van Gool. Repaint: Inpaint- ing using denoising diffusion probabilistic models.arXiv preprint arXiv:2201.09865, 2022. 2

    work page arXiv 2022

  17. [17]

    Latent Consistency Models: Synthesizing High- Resolution Images with Few-Step Inference, 2023

    Simian Luo, Yiqin Tan, Longbo Huang, Jian Li, and Hang Zhao. Latent Consistency Models: Synthesizing High- Resolution Images with Few-Step Inference, 2023. 6

    2023

  18. [18]

    The monk skin tone scale

    Ellis Monk. The monk skin tone scale. 2023. 4

    2023

  19. [19]

    T2I-Adapter: Learning Adapters to Dig out More Controllable Ability for Text-to-Image Diffusion Models

    Chong Mou, Xintao Wang, Liangbin Xie, Jian Zhang, Zhon- gang Qi, Ying Shan, and Xiaohu Feng. T2i-adapter: Learn- ing adapters to dig out more controllable ability for text-to- image diffusion models.arXiv preprint arXiv:2302.08453,

    work page internal anchor Pith review Pith/arXiv arXiv

  20. [20]

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

    Thao Nguyen, Anh Tran, and Minh Hoai. Lipstick ain’t enough: Beyond color matching for in-the-wild makeup transfer. InProceedings of the IEEE Conference on Com- puter Vision and Pattern Recognition (CVPR), 2021. 2, 5, 7

    2021

  21. [21]

    Maxime Oquab, Timoth ´ee Darcet, Theo Moutakanni, Huy V . V o, Marc Szafraniec, Vasil Khalidov, Pierre Fernandez, Daniel Haziza, Francisco Massa, Alaaeldin El-Nouby, Rus- sell Howes, Po-Yao Huang, Hu Xu, Vasu Sharma, Shang- Wen Li, Wojciech Galuba, Mike Rabbat, Mido Assran, Nico- las Ballas, Gabriel Synnaeve, Ishan Misra, Herve Jegou, Julien Mairal, Patri...

    2023

  22. [22]

    Supervised makeup transfer with a cu- rated dataset: Decoupling identity and makeup features for enhanced transformation.arXiv preprint arXiv:2602.00729,

    Qihe Pan, Yiming Wu, Xing Zhao, Liang Xie, Guodao Sun, and Ronghua Liang. Supervised makeup transfer with a cu- rated dataset: Decoupling identity and makeup features for enhanced transformation.arXiv preprint arXiv:2602.00729,

    work page arXiv

  23. [23]

    Dreammakeup: Face makeup customization using latent diffusion models.arXiv preprint arXiv:2510.10918, 2025

    Geon Yeong Park, Inhwa Han, Serin Yang, Yeobin Hong, Seongmin Jeong, Heechan Jeon, Myeongjin Goh, Sung Won Yi, Jin Nam, and Jong Chul Ye. Dreammakeup: Face makeup customization using latent diffusion models.arXiv preprint arXiv:2510.10918, 2025. 2

    work page arXiv 2025

  24. [24]

    Sdxl: Improving latent diffusion models for high-resolution image synthesis

    Dustin Podell, Zion English, Kyle Lacey, Andreas Blattmann, Tim Dockhorn, Jonas M ¨uller, Joe Penna, and Robin Rombach. Sdxl: Improving latent diffusion models for high-resolution image synthesis. InThe Twelfth Interna- tional Conference on Learning Representations, 2023. 2

    2023

  25. [25]

    Ramesh, Gabriel Goh, Sandhini Agarwal, Girish Sastry, Amanda Askell, Pamela Mishkin, Jack Clark, Gretchen Krueger, and Ilya Sutskever

    Alec Radford, Jong Wook Kim, Chris Hallacy, A. Ramesh, Gabriel Goh, Sandhini Agarwal, Girish Sastry, Amanda Askell, Pamela Mishkin, Jack Clark, Gretchen Krueger, and Ilya Sutskever. Learning transferable visual models from nat- ural language supervision. InICML, 2021. 4

    2021

  26. [26]

    Hierarchical Text-Conditional Image Generation with CLIP Latents

    Aditya Ramesh, Prafulla Dhariwal, Alex Nichol, Casey Chu, and Mark Chen. Hierarchical text-conditional image gener- ation with clip latents.arXiv preprint arXiv:2204.06125, 1 (2):3, 2022. 2

    work page internal anchor Pith review Pith/arXiv arXiv 2022

  27. [27]

    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. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 10684–10695, 2022. 2

    2022

  28. [28]

    Photorealistic text-to-image diffusion models with deep language understanding.Advances in neural information processing systems, 35:36479–36494, 2022

    Chitwan Saharia, William Chan, Saurabh Saxena, Lala Li, Jay Whang, Emily L Denton, Kamyar Ghasemipour, Raphael Gontijo Lopes, Burcu Karagol Ayan, Tim Salimans, et al. Photorealistic text-to-image diffusion models with deep language understanding.Advances in neural information processing systems, 35:36479–36494, 2022. 2

    2022

  29. [29]

    Boosted LightFace: A Hybrid DNN and GBM Model for Boosted Facial Recog- nition.Gazi University Journal of Science, 2026

    Sefik Ilkin Serengil and Alper Ozpinar. Boosted LightFace: A Hybrid DNN and GBM Model for Boosted Facial Recog- nition.Gazi University Journal of Science, 2026. 5

    2026

  30. [30]

    Shmt: Self- supervised hierarchical makeup transfer via latent diffusion models.Advances in neural information processing systems,

    Zhaoyang Sun, Shengwu Xiong, Yaxiong Chen, Fei Du, Weihua Chen, Fang Wang, and Yi Rong. Shmt: Self- supervised hierarchical makeup transfer via latent diffusion models.Advances in neural information processing systems,

  31. [31]

    Kling-Omni Technical Report

    Kling Team. Kling-omni: A generalist generative frame- work for multimodal video synthesis.arXiv preprint arXiv:2512.16776, 2025. 2

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  32. [32]

    Wan: Open and advanced large-scale video gen- erative models.arXiv preprint arXiv:2406.09203, 2024

    Wan Team. Wan: Open and advanced large-scale video gen- erative models.arXiv preprint arXiv:2406.09203, 2024. 2

    work page arXiv 2024

  33. [33]

    Wai-Shun Tong, Chi-Keung Tang, M. S. Brown, and Ying- Qing Xu. Example-based cosmetic transfer.15th Pa- cific Conference on Computer Graphics and Applications (PG’07), pages 211–218, 2007. 1, 2

    2007

  34. [34]

    Ledits: Real image editing with ddpm inversion and semantic guidance.arXiv preprint arXiv:2307.00522, 2023

    Linoy Tsaban and Apolin ´ario Passos. Ledits: Real image editing with ddpm inversion and semantic guidance.arXiv preprint arXiv:2307.00522, 2023. 2

    work page arXiv 2023

  35. [35]

    Diffusion model alignment using direct preference optimization

    Bram Wallace et al. Diffusion model alignment using direct preference optimization. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2024. 2

    2024

  36. [36]

    Unleashing high- quality image generation in diffusion sampling using second- order levenberg-marquardt-langevin

    Fangyikang Wang, Hubery Yin, Lei Qian, Yinan Li, Shaobin Zhuang, Huminhao Zhu, Yilin Zhang, Yanlong Tang, Chao Zhang, Hanbin Zhao, et al. Unleashing high- quality image generation in diffusion sampling using second- order levenberg-marquardt-langevin. InProceedings of the IEEE/CVF International Conference on Computer Vision,

  37. [37]

    InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuning

    Yichi Wu, Hongming Zhang, Xiaohui Li, Nian Wen, Yingxue Gao, Yelong Shen, and Nan Duan. A unified frame- work for incorporating human feedback into text-to-image generation.arXiv preprint arXiv:2305.06500, 2023. 2

    work page internal anchor Pith review Pith/arXiv arXiv 2023

  38. [38]

    Ramgan: Region attentive morph- ing gan for region-level makeup transfer

    Jianfeng Xiang, Junliang Chen, Wenshuang Liu, Xianxu Hou, and Linlin Shen. Ramgan: Region attentive morph- ing gan for region-level makeup transfer. InComputer Vi- sion – ECCV 2022: 17th European Conference, Tel Aviv, Israel, October 23–27, 2022, Proceedings, Part XXII, page 719–735, Berlin, Heidelberg, 2022. Springer-Verlag. 2

    2022

  39. [39]

    Edit every- thing: A text-guided generative system for images editing

    Defeng Xie, Ruichen Wang, Jian Ma, Chen Chen, Haonan Lu, Dong Yang, Fobo Shi, and Xiaodong Lin. Edit every- thing: A text-guided generative system for images editing. arXiv preprint arXiv:2304.14006, 2023. 2

    work page arXiv 2023

  40. [40]

    An automatic framework for example-based virtual makeup

    Lin Xu, Yangzhou Du, and Yimin Zhang. An automatic framework for example-based virtual makeup. InProceed- ings of the 20th IEEE International Conference on Image Processing, pages 3206–3210, 2013. 1, 2

    2013

  41. [41]

    Elegant: Exquisite and locally editable gan for makeup trans- fer

    Chenyu Yang, Wanrong He, Yingqing Xu, and Yang Gao. Elegant: Exquisite and locally editable gan for makeup trans- fer. InEuropean conference on computer vision, pages 737–

  42. [42]

    Springer, 2022. 1, 2

    2022

  43. [43]

    Depth anything: Unleashing the power of large-scale unlabeled data

    Lihe Yang, Bingyi Kang, Zilong Huang, Xiaogang Xu, Jiashi Feng, and Hengshuang Zhao. Depth anything: Unleashing the power of large-scale unlabeled data. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 10371–10381, 2024. 2, 3, 4

    2024

  44. [44]

    Lora: Low-rank adaptation for fast text-to-image diffusion fine- tuning.arXiv preprint arXiv:2307.02904, 2023

    Haonan Yu, Xiangyu Chen, Kunhao Chen, Weiwei Shi, Xi- aodong Xie, Yong Zhang, Tao Qin, and Tie-Yan Liu. Lora: Low-rank adaptation for fast text-to-image diffusion fine- tuning.arXiv preprint arXiv:2307.02904, 2023. 2

    work page arXiv 2023

  45. [45]

    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 Proceedings of the IEEE/CVF international conference on computer vision, pages 3836–3847, 2023. 2, 3, 4

    2023

  46. [46]

    Stablemakeup: When real-world makeup transfer meets dif- fusion model

    Yuxuan Zhang, Yirui Yuan, Yiren Song, and Jiaming Liu. Stablemakeup: When real-world makeup transfer meets dif- fusion model. InProceedings of the Special Interest Group on Computer Graphics and Interactive Techniques Confer- ence Conference Papers, New York, NY , USA, 2025. Asso- ciation for Computing Machinery. 1, 2, 3, 5, 7

    2025

  47. [47]

    Sine: Single image editing with text- to-image diffusion models

    Zhixing Zhang, Ligong Han, Arnab Ghosh, Dimitris N Metaxas, and Jian Ren. Sine: Single image editing with text- to-image diffusion models. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 6027–6037, 2023. 2

    2023

  48. [48]

    Flux-makeup: High-fidelity, identity- consistent, and robust makeup transfer via diffusion trans- former.arXiv preprint arXiv:2508.05069, 2025

    Jian Zhu, Shanyuan Liu, Liuzhuozheng Li, Yue Gong, He Wang, Bo Cheng, Yuhang Ma, Liebucha Wu, Xiaoyu Wu, Dawei Leng, et al. Flux-makeup: High-fidelity, identity- consistent, and robust makeup transfer via diffusion trans- former.arXiv preprint arXiv:2508.05069, 2025. 2, 3

    work page arXiv 2025