pith. sign in

arxiv: 2606.29319 · v1 · pith:OOTRB37Unew · submitted 2026-06-28 · 💻 cs.CV

FDM-MFVT: Few-step Sampling Diffusion Model for Mask-Free Virtual Try-On

Jiaxin Liu , Xiaoye Liang , Lai Jiang , Mai Xu , Jun Liu This is my paper

Pith reviewed 2026-06-30 07:40 UTC · model grok-4.3

classification 💻 cs.CV
keywords virtual try-ondiffusion modelsmask-freefew-step samplingimage synthesisnoise optimizationinstruction-driven adaptationMFVT dataset
0
0 comments X

The pith

A diffusion model produces mask-free virtual try-on images in six sampling steps by optimizing noise from the input image alone.

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

The paper presents a diffusion model for image-based virtual try-on that removes the requirement for clothing masks and the auxiliary networks that generate them. It introduces an outfit-aware noise optimization step that starts the diffusion process directly from noise extracted from the person image, reaching usable results after only six steps rather than the usual thirty. An instruction-driven module then adapts the output using text prompts that describe the desired try-on, and the work supplies a new dataset of thirty thousand mask-free image pairs to train and evaluate such systems. If the approach holds, virtual try-on would require less computation and fewer preprocessing stages while still matching or exceeding the quality of mask-dependent methods.

Core claim

FDM-MFVT combines an Outfit-aware Noise Optimization Module (OANO) that initializes the alignment space with noise taken from the input person image and an Instruction-driven Try-on Module (IDT) that performs efficient adaptation guided by virtual try-on prompts. Together these components generate higher-fidelity try-on images from garment and person images alone after only six diffusion steps, outperforming both mask-based and mask-free baselines that need more steps, while the accompanying MFVT dataset of thirty thousand pairs removes the previous data bottleneck for mask-free training.

What carries the argument

The Outfit-aware Noise Optimization (OANO) module, which derives an initial noise field directly from the input image to set up the diffusion alignment space for few-step generation.

If this is right

  • Only six diffusion steps are needed instead of thirty, lowering the compute required per try-on output.
  • No masks or separate mask-prediction networks are required at inference time.
  • The IDT module produces results from garment and person images alone when guided by text prompts.
  • A new thirty-thousand-pair mask-free dataset is released to support further work on this task.
  • Quantitative and qualitative metrics exceed those of existing mask-based and mask-free baselines.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Mobile or web applications could run virtual try-on in real time because the step count drops to six.
  • The same noise-initialization idea might apply to other diffusion-based image-editing tasks that currently depend on masks.
  • Performance on body shapes or clothing categories underrepresented in the new dataset would indicate how far the input-image noise initialization generalizes.
  • E-commerce sites could integrate the method without first running a separate segmentation model on every user photo.

Load-bearing premise

Noise taken from the input image alone is sufficient to create an alignment space that yields high-fidelity try-on results after only six diffusion steps without any masks.

What would settle it

A side-by-side comparison on person images with complex poses or garments whose texture and fit differ sharply from the target clothing, checking whether visible misalignment or artifacts remain after exactly six steps.

Figures

Figures reproduced from arXiv: 2606.29319 by Jiaxin Liu, Jun Liu, Lai Jiang, Mai Xu, Xiaoye Liang.

Figure 1
Figure 1. Figure 1: We propose FDM-MFVT, a mask-free few-step virtual try-on framework that achieves high-fidelity virtual try-on with fewer inference steps across multiple scenarios and requires no reference conditions. Abstract. Image-based Virtual Try-On (IVTON) has greatly advanced through diffusion models, yet existing methods require many sampling steps and depend on masks with costly auxiliary networks. In addition, th… view at source ↗
Figure 2
Figure 2. Figure 2: Illustration of the FDM-MFVT framework. The left part shows the Outfit￾aware Noise Optimization (OANO) module, which refines noise initialization. The right part shows the Instruction-driven Try-on (IDT) module, which integrates visual and textual features for mask-free synthesis. while enabling high-quality outputs with minimal steps. Building on the opti￾mized noise x ′ T , the IDT module integrates both… view at source ↗
Figure 3
Figure 3. Figure 3: Construction pipeline of the MFVT dataset. In the mask-based setting, we compare our method against eight state-of-the￾art methods: StableVTON [24], OOTDiffusion [43], CATVTON [6], TPD [43], IDM-VITON [5], ITA-MDT [20], OmniVTON [45] and FastFit [7]. We report both quantitative and qualitative results. Specifically, Quantitative evaluation includes objective metrics such as LPIPS [46], SSIM [42], FID [19],… view at source ↗
Figure 4
Figure 4. Figure 4: Qualitative results of our method on the StreetVTON dataset. 4.3 Qualitative Results Here, we present a comprehensive comparison of qualitative results between our method and both mask-free and mask-based baselines, focusing on garment fi￾delity, texture realism, and overall visual coherence. Furthermore, we validated the scenario generalization ability of our method in the wild dataset as shown in [PITH_… view at source ↗
Figure 5
Figure 5. Figure 5: Qualitative comparison between our method and Any2AnyTryon on the MFVT dataset. Diffusion [43], CATVTON [6], TPD [43], IDM-VITON [5], ITA-MDT [20], Om￾niVTON [45] and FastFit [7]. All evaluations are performed on the unpaired test sets of the VTON-HD [4] and DressCode [32] datasets. As illustrated in [PITH_FULL_IMAGE:figures/full_fig_p012_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Qualitative comparison between our method and eihgt baselines on the unpaired test sets of VTON-HD [4] and DressCode [32] Upper-body datasets. leads to a noticeable degradation in perceptual quality, with LPIPS increasing from 0.091 to 0.121 and SSIM dropping from 0.879 to 0.864. This confirms the importance of noise refinement in enhancing structural consistency and visual realism. Similarly, excluding ID… view at source ↗
Figure 7
Figure 7. Figure 7: Qualitative results of FDM-MFVT ablation. Efficiency Analysis The OANO module significantly reduced the number of inference steps and obtained higher fidelity virtual try-on images in just 0.28s. In addition, the OANO module has a small number of parameters and is easy to train. These factors highlight the lightweight, efficient, and broad application potential of OANO module. 5 Conclusion In this work, we… view at source ↗
read the original abstract

Image-based Virtual Try-On (IVTON) has greatly advanced through diffusion models, yet existing methods require many sampling steps and depend on masks with costly auxiliary networks. In addition, the absence of large-scale mask-free paired datasets further limits the development of mask-free IVTON. We propose FDM-MFVT, a few-step diffusion model for mask-free IVTON, integrating an Outfit-aware Noise Optimization Module (OANO) and an Instruction-driven Try-on Module (IDT) to enhance efficiency and flexibility.The OANO module initializes the alignment space with noise using the input image and only needs 6 steps to generate a higher-fidelity try-on image compared to 30 steps.The IDT module uses virtual try-on prompts and efficient adaptation to generate high-quality results from garment and person images alone. We further introduce MFVT, a 30,000-pair mask-free IVTON dataset. Experiments show that FDM-MFVT achieves superior quantitative and qualitative results with fewer inference steps than mask-based and mask-free baseline methods.

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

2 major / 0 minor

Summary. The paper proposes FDM-MFVT, a few-step diffusion model for mask-free image-based virtual try-on. It introduces an Outfit-aware Noise Optimization (OANO) module that initializes the alignment space using noise derived from the input image alone, an Instruction-driven Try-on (IDT) module that uses virtual try-on prompts for generation from garment and person images, and a new MFVT dataset of 30,000 mask-free pairs. The central claim is that the method produces higher-fidelity outputs in only 6 diffusion steps, outperforming both mask-based and mask-free baselines that require 30 steps.

Significance. If the empirical claims hold, the work would advance practical IVTON by demonstrating that mask-free generation is feasible with substantially reduced sampling steps and without auxiliary networks, addressing efficiency and data limitations in the field. The release of the MFVT dataset would also provide a concrete resource for future mask-free research.

major comments (2)
  1. [Abstract] Abstract: The assertion of 'superior quantitative and qualitative results' with 6 steps is presented without any reported metrics (e.g., FID, LPIPS, SSIM), baseline comparisons, ablation studies, or error analysis, which is load-bearing for the central claim of outperformance over mask-based and mask-free methods.
  2. [Abstract] Abstract (OANO description): No information is supplied on the optimization objective inside OANO, the distribution of training poses/garments, or how noise derived from the input image alone produces reliable garment-person alignment; this initialization is the load-bearing assumption for the 6-step inference claim without masks or extra networks.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their review and the opportunity to clarify points regarding the abstract. We address each major comment below with references to the full manuscript.

read point-by-point responses

  1. Referee: [Abstract] Abstract: The assertion of 'superior quantitative and qualitative results' with 6 steps is presented without any reported metrics (e.g., FID, LPIPS, SSIM), baseline comparisons, ablation studies, or error analysis, which is load-bearing for the central claim of outperformance over mask-based and mask-free methods.

    Authors: The abstract provides a high-level summary of the contributions and claims. Detailed quantitative results including FID, LPIPS, and SSIM metrics, direct comparisons to mask-based and mask-free baselines (all using 30 steps), ablation studies on OANO and IDT, and error analysis are reported in Section 4 (Experiments) of the manuscript, where FDM-MFVT is shown to outperform baselines at 6 steps. We can revise the abstract to incorporate key numerical results if the editor prefers a more detailed summary. revision: partial

  2. Referee: [Abstract] Abstract (OANO description): No information is supplied on the optimization objective inside OANO, the distribution of training poses/garments, or how noise derived from the input image alone produces reliable garment-person alignment; this initialization is the load-bearing assumption for the 6-step inference claim without masks or extra networks.

    Authors: The optimization objective for OANO (a combined reconstruction and feature alignment loss), the MFVT dataset statistics (30,000 pairs with diverse poses, garments, and body types), and the alignment mechanism (initializing the diffusion latent space from person-image noise to encode identity and pose priors for subsequent prompt-guided garment synthesis via IDT) are fully specified in Section 3.1. These elements enable the mask-free 6-step inference without auxiliary networks, as validated in the experiments. revision: no

Circularity Check

0 steps flagged

No circularity in derivation chain

full rationale

The paper proposes FDM-MFVT by introducing two new modules (OANO for noise initialization and IDT for prompt-driven try-on) plus a new 30k-pair dataset MFVT. The abstract and provided text contain no equations, no fitted parameters renamed as predictions, no self-citation chains invoked as uniqueness theorems, and no ansatzes smuggled via prior work. All performance claims rest on empirical comparison to baselines rather than any reduction of outputs to inputs by construction. The derivation is therefore self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 3 invented entities

Review performed on abstract only; no explicit free parameters, axioms, or invented entities beyond the high-level module and dataset names are stated.

axioms (1)
  • domain assumption Diffusion models can be conditioned on garment and person images for virtual try-on without explicit masks.
    Core premise of the proposed method, standard in recent IVTON literature but unverified here.
invented entities (3)
  • OANO module no independent evidence
    purpose: Initialize alignment space with input-image noise for few-step sampling
    New component introduced to achieve 6-step generation.
  • IDT module no independent evidence
    purpose: Generate try-on results from garment and person images using prompts
    New component for mask-free operation.
  • MFVT dataset no independent evidence
    purpose: Provide 30,000 mask-free paired examples for training
    New dataset introduced to address data limitation.

pith-pipeline@v0.9.1-grok · 5716 in / 1377 out tokens · 28979 ms · 2026-06-30T07:40:24.641208+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

47 extracted references · 11 canonical work pages · 3 internal anchors

  1. [1]

    In: Proceedings of the Computer Vision and Pattern Recognition Conference

    Avrahami, O., Patashnik, O., Fried, O., Nemchinov, E., Aberman, K., Lischin- ski, D., Cohen-Or, D.: Stable flow: Vital layers for training-free image editing. In: Proceedings of the Computer Vision and Pattern Recognition Conference. pp. 7877–7888 (2025)

    2025

  2. [2]

    arXiv preprint arXiv:2303.04248 , year=

    Berthelot,D.,Autef,A.,Lin,J.,Yap,D.A.,Zhai,S.,Hu,S.,Zheng,D.,Talbott,W., Gu, E.: Tract: Denoising diffusion models with transitive closure time-distillation. arXiv preprint arXiv:2303.04248 (2023)

    work page arXiv 2023

  3. [3]

    In: Pro- ceedings of the IEEE/CVF international conference on computer vision

    Cao, M., Wang, X., Qi, Z., Shan, Y., Qie, X., Zheng, Y.: Masactrl: Tuning-free mutual self-attention control for consistent image synthesis and editing. In: Pro- ceedings of the IEEE/CVF international conference on computer vision. pp. 22560– 22570 (2023)

    2023

  4. [4]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Choi, S., Park, S., Lee, M., Choo, J.: Viton-hd: High-resolution virtual try-on via misalignment-aware normalization. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 14131–14140 (2021)

    2021

  5. [5]

    In: European Conference on Computer Vision

    Choi, Y., Kwak, S., Lee, K., Choi, H., Shin, J.: Improving diffusion models for authentic virtual try-on in the wild. In: European Conference on Computer Vision. pp. 206–235. Springer (2024)

    2024

  6. [6]

    arXiv preprint arXiv:2407.15886 (2024)

    Chong, Z., Dong, X., Li, H., Zhang, S., Zhang, W., Zhang, X., Zhao, H., Jiang, D., Liang, X.: Catvton: Concatenation is all you need for virtual try-on with diffusion models. arXiv preprint arXiv:2407.15886 (2024)

    work page arXiv 2024

  7. [7]

    arXiv preprint arXiv:2508.20586 (2025)

    Chong, Z., Lei, Y., Zhang, S., He, Z., Wang, Z., Zhang, X., Dong, X., Wu, Y., Jiang, D., Liang, X.: Fastfit: Accelerating multi-reference virtual try-on via cacheable diffusion models. arXiv preprint arXiv:2508.20586 (2025)

    work page arXiv 2025

  8. [8]

    In: Pro- ceedings of the Winter Conference on Applications of Computer Vision

    Cui, A., Mahajan, J., Shah, V., Gomathinayagam, P., Liu, C., Lazebnik, S.: Street tryon: Learning in-the-wild virtual try-on from unpaired person images. In: Pro- ceedings of the Winter Conference on Applications of Computer Vision. pp. 1414– 1423 (2025)

    2025

  9. [9]

    ACM Transactions on Multimedia Computing, Com- munications and Applications19(1s), 1–21 (2023)

    De Divitiis, L., Becattini, F., Baecchi, C., Del Bimbo, A.: Disentangling features for fashion recommendation. ACM Transactions on Multimedia Computing, Com- munications and Applications19(1s), 1–21 (2023)

    2023

  10. [10]

    arXiv preprint arXiv:2508.13632 (2025)

    Feng, Y., Zhang, L., Cao, H., Chen, Y., Feng, X., Cao, J., Wu, Y., Wang, B.: Omnitry: Virtual try-on anything without masks. arXiv preprint arXiv:2508.13632 (2025)

    work page arXiv 2025

  11. [11]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Ge, Y., Song, Y., Zhang, R., Ge, C., Liu, W., Luo, P.: Parser-free virtual try-on via distilling appearance flows. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 8485–8493 (2021)

    2021

  12. [12]

    In: ICML 2023 Workshop on Structured Probabilistic Inference{\&}Generative Modeling

    Gu, J., Zhai, S., Zhang, Y., Liu, L., Susskind, J.M.: Boot: Data-free distillation of denoising diffusion models with bootstrapping. In: ICML 2023 Workshop on Structured Probabilistic Inference{\&}Generative Modeling. vol. 3 (2023)

    2023

  13. [13]

    In: Proceedings of the IEEE conference on computer vision and pattern recognition

    Güler, R.A., Neverova, N., Kokkinos, I.: Densepose: Dense human pose estimation in the wild. In: Proceedings of the IEEE conference on computer vision and pattern recognition. pp. 7297–7306 (2018) 16 J, Liu et al

    2018

  14. [14]

    In: Proceed- ings of the IEEE/CVF International Conference on Computer Vision

    Guo,H.,Zeng,B.,Song,Y.,Zhang,W.,Liu,J.,Zhang,C.:Any2anytryon:Leverag- ing adaptive position embeddings for versatile virtual clothing tasks. In: Proceed- ings of the IEEE/CVF International Conference on Computer Vision. pp. 19085– 19096 (2025)

    2025

  15. [15]

    In: Proceedings of the IEEE international conference on computer vision

    Hadi Kiapour, M., Han, X., Lazebnik, S., Berg, A.C., Berg, T.L.: Where to buy it: Matching street clothing photos in online shops. In: Proceedings of the IEEE international conference on computer vision. pp. 3343–3351 (2015)

    2015

  16. [16]

    In: Proceedings of the IEEE/CVF international conference on computer vision

    Han,X.,Hu,X.,Huang,W.,Scott,M.R.:Clothflow:Aflow-basedmodelforclothed person generation. In: Proceedings of the IEEE/CVF international conference on computer vision. pp. 10471–10480 (2019)

    2019

  17. [17]

    In: Proceedings of the IEEE conference on computer vision and pattern recognition

    Han, X., Wu, Z., Wu, Z., Yu, R., Davis, L.S.: Viton: An image-based virtual try-on network. In: Proceedings of the IEEE conference on computer vision and pattern recognition. pp. 7543–7552 (2018)

    2018

  18. [18]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    He, S., Song, Y.Z., Xiang, T.: Style-based global appearance flow for virtual try- on. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 3470–3479 (2022)

    2022

  19. [19]

    Advances in neural information processing systems30(2017)

    Heusel,M.,Ramsauer,H.,Unterthiner,T.,Nessler,B.,Hochreiter,S.:Ganstrained by a two time-scale update rule converge to a local nash equilibrium. Advances in neural information processing systems30(2017)

    2017

  20. [20]

    In: Proceedings of the Computer Vision and Pattern Recognition Confer- ence

    Hong, J.W., Ton, T., Pham, T.X., Koo, G., Yoon, S., Yoo, C.D.: Ita-mdt: Image- timestep-adaptive masked diffusion transformer framework for image-based virtual try-on. In: Proceedings of the Computer Vision and Pattern Recognition Confer- ence. pp. 28284–28294 (2025)

    2025

  21. [21]

    In: Proceedings of the IEEE conference on computer vision and pattern recognition

    Hsiao, W.L., Grauman, K.: Creating capsule wardrobes from fashion images. In: Proceedings of the IEEE conference on computer vision and pattern recognition. pp. 7161–7170 (2018)

    2018

  22. [22]

    In-context lora for diffusion transformers.arXiv preprint arXiv:2410.23775, 2024

    Huang,L.,Wang,W.,Wu,Z.F.,Shi,Y.,Dou,H.,Liang,C.,Feng,Y.,Liu,Y.,Zhou, J.: In-context lora for diffusion transformers. arXiv preprint arXiv:2410.23775 (2024)

    work page arXiv 2024

  23. [23]

    In: European Conference on Computer Vision

    Issenhuth, T., Mary, J., Calauzenes, C.: Do not mask what you do not need to mask: a parser-free virtual try-on. In: European Conference on Computer Vision. pp. 619–635. Springer (2020)

    2020

  24. [24]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Kim, J., Gu, G., Park, M., Park, S., Choo, J.: Stableviton: Learning semantic correspondence with latent diffusion model for virtual try-on. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 8176– 8185 (2024)

    2024

  25. [25]

    arXiv preprint arXiv:1907.10830 (2019)

    Kim, J., Kim, M., Kang, H., Lee, K.: U-gat-it: Unsupervised generative attentional networks with adaptive layer-instance normalization for image-to-image transla- tion. arXiv preprint arXiv:1907.10830 (2019)

    work page arXiv 1907

  26. [26]

    In: European Conference on Com- puter Vision

    Lee, S., Gu, G., Park, S., Choi, S., Choo, J.: High-resolution virtual try-on with misalignment and occlusion-handled conditions. In: European Conference on Com- puter Vision. pp. 204–219. Springer (2022)

    2022

  27. [27]

    Advances in Neural Information Processing Systems36, 20662–20678 (2023)

    Li, Y., Wang, H., Jin, Q., Hu, J., Chemerys, P., Fu, Y., Wang, Y., Tulyakov, S., Ren, J.: Snapfusion: Text-to-image diffusion model on mobile devices within two seconds. Advances in Neural Information Processing Systems36, 20662–20678 (2023)

    2023

  28. [28]

    VTEdit-Bench: A Comprehensive Benchmark for Multi-Reference Image Editing Models in Virtual Try-On

    Liang, X., Qu, Z., Zou, M., Liu, J., Jiang, L., Xu, M., Zhu, Y.: Vtedit-bench: A comprehensive benchmark for multi-reference image editing models in virtual try-on. arXiv preprint arXiv:2603.11734 (2026) FDM-MFVT 17

    work page internal anchor Pith review Pith/arXiv arXiv 2026

  29. [29]

    SDEdit: Guided Image Synthesis and Editing with Stochastic Differential Equations

    Meng, C., He, Y., Song, Y., Song, J., Wu, J., Zhu, J.Y., Ermon, S.: Sdedit: Guided image synthesis and editing with stochastic differential equations. arXiv preprint arXiv:2108.01073 (2021)

    work page internal anchor Pith review Pith/arXiv arXiv 2021

  30. [30]

    In: Proceedings of the IEEE/CVF confer- ence on computer vision and pattern recognition

    Meng, C., Rombach, R., Gao, R., Kingma, D., Ermon, S., Ho, J., Salimans, T.: On distillation of guided diffusion models. In: Proceedings of the IEEE/CVF confer- ence on computer vision and pattern recognition. pp. 14297–14306 (2023)

    2023

  31. [31]

    In: CVPR workshops

    Minar, M.R., Tuan, T.T., Ahn, H., Rosin, P., Lai, Y.K.: Cp-vton+: Clothing shape and texture preserving image-based virtual try-on. In: CVPR workshops. vol. 3, pp. 10–14 (2020)

    2020

  32. [32]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Morelli, D., Fincato, M., Cornia, M., Landi, F., Cesari, F., Cucchiara, R.: Dress code: High-resolution multi-category virtual try-on. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 2231–2235 (2022)

    2022

  33. [33]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Ning, S., Wang, D., Qin, Y., Jin, Z., Wang, B., Han, X.: Picture: Photorealis- tic virtual try-on from unconstrained designs. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 6976–6985 (2024)

    2024

  34. [34]

    arXiv preprint arXiv:2407.14041 (2024)

    Qi, Z., Bai, L., Xiong, H., Xie, Z.: Not all noises are created equally: Diffusion noise selection and optimization. arXiv preprint arXiv:2407.14041 (2024)

    work page arXiv 2024

  35. [35]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

    Rombach, R., Blattmann, A., Lorenz, D., Esser, P., Ommer, B.: High-resolution image synthesis with latent diffusion models. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). pp. 10684– 10695 (2022)

    2022

  36. [36]

    Progressive Distillation for Fast Sampling of Diffusion Models

    Salimans, T., Ho, J.: Progressive distillation for fast sampling of diffusion models. arXiv preprint arXiv:2202.00512 (2022)

    work page internal anchor Pith review Pith/arXiv arXiv 2022

  37. [37]

    In: Proceedings of the IEEE/CVF winter conference on applications of computer vision

    Sarkar, R., Bodla, N., Vasileva, M.I., Lin, Y.L., Beniwal, A., Lu, A., Medioni, G.: Outfittransformer: Learning outfit representations for fashion recommendation. In: Proceedings of the IEEE/CVF winter conference on applications of computer vision. pp. 3601–3609 (2023)

    2023

  38. [38]

    Song, Y., Dhariwal, P., Chen, M., Sutskever, I.: Consistency models (2023)

    2023

  39. [39]

    In: Proceed- ings of the International Conference on Machine Learning (ICML)

    Song, Y., Meng, C., Ermon, S.: Denoising diffusion implicit models. In: Proceed- ings of the International Conference on Machine Learning (ICML). pp. 9198–9207 (2020)

    2020

  40. [40]

    IEEE Transactions on Image Processing27(12), 6283–6294 (2018)

    Wang, B., Zheng, H., Liang, X., Shen, X., Shao, L.: Toward photo-realistic virtual try-on by adaptively generating-preserving image content. IEEE Transactions on Image Processing27(12), 6283–6294 (2018)

    2018

  41. [41]

    Mod- eling and predicting single-cell multi-gene perturbation responses with scLAMBDA.bioRxiv, 2024a

    Wang, J., Pu, J., Qi, Z., Guo, J., Ma, Y., Huang, N., Chen, Y., Li, X., Shan, Y.: Taming rectified flow for inversion and editing. arXiv preprint arXiv:2411.04746 (2024)

    work page arXiv 2024

  42. [42]

    IEEE transactions on image processing 13(4), 600–612 (2004)

    Wang, Z., Bovik, A.C., Sheikh, H.R., Simoncelli, E.P.: Image quality assessment: from error visibility to structural similarity. IEEE transactions on image processing 13(4), 600–612 (2004)

    2004

  43. [43]

    In: Proceedings of the AAAI Conference on Artificial Intelligence

    Xu, Y., Gu, T., Chen, W., Chen, A.: Ootdiffusion: Outfitting fusion based latent diffusion for controllable virtual try-on. In: Proceedings of the AAAI Conference on Artificial Intelligence. vol. 39, pp. 8996–9004 (2025)

    2025

  44. [44]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Yang, X., Ding, C., Hong, Z., Huang, J., Tao, J., Xu, X.: Texture-preserving dif- fusion models for high-fidelity virtual try-on. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 7017–7026 (2024)

    2024

  45. [45]

    In: Proceedings of the IEEE/CVF International Confer- ence on Computer Vision

    Yang, Z., Li, Y., He, S., Li, X., Xu, Y., Dong, J., Du, Y.: Omnivton: Training-free universal virtual try-on. In: Proceedings of the IEEE/CVF International Confer- ence on Computer Vision. pp. 16702–16711 (2025) 18 J, Liu et al

    2025

  46. [46]

    In: Proceedings of the IEEE conference on computer vision and pattern recognition

    Zhang, R., Isola, P., Efros, A.A., Shechtman, E., Wang, O.: The unreasonable effectiveness of deep features as a perceptual metric. In: Proceedings of the IEEE conference on computer vision and pattern recognition. pp. 586–595 (2018)

    2018

  47. [47]

    In: Proceedings of the IEEE/CVF International Conference on Computer Vision

    Zhou, Z., Shao, S., Bai, L., Zhang, S., Xu, Z., Han, B., Xie, Z.: Golden noise for diffusion models: A learning framework. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 17688–17697 (2025)

    2025