Pith. sign in

REVIEW 5 major objections 6 minor 35 references

Real-Time Person Image Synthesis Using a Flow Matching Model

T0 review · 5 major / 6 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read Flow matching makes pose-guided person image synthesis near real-time.

desk verdict First flow-matching application to pose-guided person synthesis with a credible speed advantage, but the batch-8 timing doesn't support the frame-by-frame real-time claim. read the letter →

arxiv 2505.03562 v1 pith:IXQ52YFD submitted 2025-05-06 cs.CV cs.AI

Jiwoo Jeong , Kirok Kim , Wooju Kim , Nam-Joon Kim This is my paper · ORCID
classification cs.CVcs.AI
keywords pose-guidedpersonimagesynthesisflowmatchinglatentspaceDiTconditionalgenerationreal-timeDeepFashionclassifier-freeguidance
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

The paper proposes RPFM, a pose-guided person image synthesis model built on flow matching instead of diffusion sampling, and argues that this swap is what makes near-real-time generation possible without a large quality penalty. It claims that on the DeepFashion benchmark, the medium model generates eight 256x256 images in 0.489 seconds on one A100 GPU, roughly 2.4 times faster than the fastest published diffusion transformer baseline, while its quality metrics stay close to that baseline's large model. If true, this would make frame-by-frame video applications like sign language synthesis, AR/VR, gaming, and live streaming practical, because the bottleneck is no longer the many denoising steps of diffusion. The design point is a deliberate trade: a small, acceptable drop in image accuracy in exchange for a more than twofold speed increase.

What carries the argument

The load-bearing object is the conditional optimal transport flow matching path, written $x_t=(1-t)x_0+t x_1$, which defines a straight-line interpolation from Gaussian noise to a data sample; the network trains to predict the velocity $rac{dx}{dt}$ and sampling solves the ODE with a small number of function evaluations. Around this, the paper builds a DiT-style transformer backbone operating on a 32x32x12 latent tensor formed by concatenating the noisy latent, the VAE-encoded source image, and the VAE-encoded target pose, and it adds a condition branch that aggregates four embeddings: local and global features of both the source image (patchified VAE tokens for local, DINOv2-G for global) and the target pose (patchified VAE tokens for local, DINOv2-B for global) into a single conditioning vector. This combination lets the model condition on both appearance and pose with minimal extra forward passes, and the paper's ablations show that removing either component degrades LPIPS, SSIM, and PSNR while keeping the other.

What would settle it

Re-run X-MDPT-S and X-MDPT-L on the same machine, with the same resizing to 256x256 and 512x512, same batching, and same FID, LPIPS, SSIM, and PSNR evaluation code, and compare latency and metrics side by side; if the diffusion baselines are also under half a second per batch or the quality gap grows beyond the paper's reported margin, the near-real-time and comparable-quality claims weaken.

Watch

Extended reading notes

Core claim

The central claim is that replacing the diffusion process in pose-guided person image synthesis with a flow matching model, run in the latent space of a pretrained VAE and backed by a DiT transformer, yields images that follow the target pose and preserve the source appearance while sampling much faster. In the paper's measurements this is concretized as 0.489 seconds for a batch of eight 256x256 images at NFE=60, versus 1.191 seconds for X-MDPT-S, with a 512x512-quality comparison showing FID 8.522, LPIPS 0.1786, SSIM 0.7742 against X-MDPT-L's 7.162, 0.1645, 0.7522. The paper also reports the highest SSIM among the compared models at 512x512 and second-best LPIPS, and argues that because flow matching uses deterministic ODE trajectories rather than stochastic denoising, the sampling path can be shortened without the instability that would come from removing diffusion noise injection steps.

Load-bearing premise

The comparison assumes that the baseline quality and speed numbers, taken from earlier papers, were measured under the same preprocessing, hardware, batching, and metric protocol as the authors' own runs, and that batched pose frames are a valid deployment mode for real-time video.

Editorial extensions

If this is right

  • At the reported settings, a single A100 can generate a batch of eight 256x256 pose-matched frames in about half a second, so a sign-language or streaming pipeline that poses frames in batches could approach real-time per-frame throughput.
  • The quality/speed trade is explicitly chosen: the paper's medium model trades a small FID increase for the speed gain, and reports higher SSIM than the diffusion baseline at 512x512.
  • Classifier-free guidance at scale 1.1 gives the best balance across FID, LPIPS, SSIM, and PSNR for both model sizes, so the method is not sensitive to tuning this scale.
  • The four-branch conditional aggregation (local and global features for both source and pose) is the version that works best; a three-branch variant inspired by X-MDPT is worse on FID despite a similar training budget.

Reading between the lines

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

  • Editorial extension: the same flow-matching formulation should transfer to other conditional image-translation tasks that currently rely on diffusion, such as depth-to-image or sketch-to-image, because the conditioning mechanisms are task-agnostic once a VAE latent and a condition encoder exist.
  • Editorial extension: the paper's speed numbers are for batched sampling, so the leap from 0.489 seconds per batch to real-time video assumes a deployment mode that renders several pose frames at once; a frame-by-frame latency measurement would be a stricter test of real-time claims.
  • Editorial extension: since flow matching paths are deterministic, the model could in principle be stepped with adaptive ODE solvers to trade NFE against quality per input, a knob the paper did not explore.
  • Editorial extension: the higher SSIM at 512x512 despite lower FID hints that the loss of global distribution fidelity and gain in pixel-level alignment may not move together; probing with perceptual metrics or user studies could clarify which error mode matters for live applications.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

5 major / 6 minor

Summary. The paper proposes RPFM, a latent-space flow matching model with a DiT backbone for pose-guided person image synthesis (PGPIS). The model is conditioned on a source image and a target pose through two complementary mechanisms: input concatenation of the noisy latent, source latent, and pose latent, and conditional aggregation of four embeddings (local and global features of both the source image and target pose). The authors train on DeepFashion, report FID, LPIPS, SSIM, and PSNR at 256x256 and 512x512, and measure inference speed on a single A100 with batch size 8. They claim quality comparable to X-MDPT and CFLD while being about 2.4x faster than X-MDPT-S, and they conclude that the model is capable of real-time frame-by-frame video generation. Ablations support the benefit of the two conditioning components, and the paper includes a code link.

Significance. If the speed and quality claims are rigorously established, RPFM would be a useful step toward real-time PGPIS for sign language and interactive applications. The paper provides a clean architectural contribution (flow matching with concatenated and aggregated conditioning), a useful comparison across several baselines, ablations of the proposed components, and a public code release. However, the central 'real-time' claim is not supported by the presented measurements, the evaluation protocol mixes baseline numbers from different papers without statistical safeguards, and the classifier-free guidance implementation is not fully specified. These issues currently prevent the paper from supporting its headline contribution.

major comments (5)
  1. [Section 4.3 and abstract] The claim that the model is 'capable of real-time video generation on a frame-by-frame basis' is not supported by the reported data. The experiment measures generation of eight 256x256 images concurrently on one A100, and the amortized per-image time is 0.489s/8 = 61ms, or about 16 FPS, which is below the 24-30 FPS typically required for real-time video. The paper never reports a batch-1 latency, nor an end-to-end per-frame time that includes VAE decoding and pose extraction. The observed decrease in amortized per-image time with increasing batch size does not establish frame-by-frame latency. Please either provide per-frame latency at batch size 1 with the full pipeline, or revise the real-time claim to a batch-processing throughput claim.
  2. [Section 3.2, Eq. (7), and Algorithm 1] The classifier-free guidance formula requires an unconditional model output v_theta(z_t,t). The paper defines this as 'the scenario involving input concatenation without the application of conditional aggregation,' but the training procedure described in Algorithm 1 and Section 4.1 never mentions condition dropout or a null-condition embedding. Without such a training mechanism, it is unclear how the unconditional output v_theta(z_t,t) is obtained at inference for the trained model, and the CFG results (including the chosen gamma=1.1) are not reproducible. Please specify how the unconditional pass is implemented and whether the model was trained with condition dropout.
  3. [Section 4.1 and Tables 1-2] The quality comparison is built on baseline numbers sourced from the CFLD and X-MDPT papers, with no re-running of those models under the authors' evaluation protocol. The paper states that results are 'assessed at a resolution of 256x176' after generation at 256x256 or 512x512, but it does not establish that the baselines used the same resizing, cropping, or FID computation pipeline. Additionally, no variance or confidence intervals are reported for FID, LPIPS, SSIM, or PSNR for any model, including the authors' own. As a result, the claim of 'comparable' quality is not statistically grounded. Please either re-run the baselines under identical protocols or report uncertainty and a detailed protocol for each baseline.
  4. [Abstract and Table 3] The headline speed comparison is presented as RPFM-Medium (0.489s) versus X-MDPT-S (1.191s), which is the smallest X-MDPT configuration. While Table 3 also includes X-MDPT-L versus RPFM-Large, the abstract's claim of 'over a twofold increase in generation speed' relies on the most favorable configuration. Moreover, all speed numbers are batch-of-8 timings on a single A100, not per-frame latencies. Please report speed for all model sizes at batch size 1 and match baseline configurations at comparable model scale, or clearly qualify the comparison.
  5. [Tables 4-5 and Section 4.1] The CFG scale is selected based on performance on what appears to be the same test set used for the main results; no validation split is mentioned. Selecting hyperparameters on the test metrics and then reporting those metrics as the final result can overstate performance. Please perform model selection on a validation set and report the final results on a held-out test set.
minor comments (6)
  1. [Abstract] The title says 'Real-Time' while the abstract says 'near-real-time'; please align the wording with what the data actually support.
  2. [Algorithm 1] Line 8 of Algorithm 1 contains a formatting error in the gradient step expression; the loss and gradient are not legible.
  3. [Tables 1-2] The NFE for the medium model differs between Table 1 (NFE=65) and Table 2 (NFE=55) without explanation; please clarify why different NFE values are used at the two resolutions.
  4. [Section 4.1] The implementation details do not specify which pretrained VAE is used (e.g., the Stable Diffusion VAE); please state the exact VAE and DINOv2 checkpoint versions.
  5. [Supplementary Figure 5] The qualitative comparison uses baseline images downloaded from the CFLD GitHub rather than generated under the same protocol; this should be explicitly stated as a non-controlled qualitative comparison.
  6. [Section 3.2] The sentence 'the flow matching model can be conceptualized as a generalized framework that includes diffusion models' is vague; please add a citation or a precise explanation.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: RPFM's flow-matching derivation and empirical speed/quality comparisons are self-contained; the real-time extrapolation issue is a correctness/logic concern, not circularity.

full rationale

The paper's technical derivation is standard conditional flow matching in latent space (Eqs. 4-6), cited to Lipman et al. and LFM, and the proposed model's outputs are evaluated with external metrics (FID, LPIPS, SSIM, PSNR) against published baselines. No fitted parameter is renamed as a prediction: CFG scale and NFE are selected by ablation/tuning, not used to force the reported metrics. The paper contains no load-bearing self-citations; references to X-MDPT, CFLD, PIDM, etc. are external prior work, and the baseline numbers are taken from those publications (a comparability limitation, not circularity). The inference from batch-size-8 timing (0.489 s for eight images) to 'real-time video generation on a frame-by-frame basis' is an unsupported extrapolation and should be flagged as a correctness risk, but it is not a derivation that reduces to its own inputs. Therefore no circular step is established.

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

The paper introduces no new physical or theoretical entities; it builds on pretrained components and existing flow matching theory. The main fitted quantities are the tuned hyperparameters (CFG scale and NFE), which are selected on the test metrics that are subsequently reported as final results.

free parameters (2)
  • CFG scale gamma = 1.1 (256x256), 1.0 (512x512)
    Chosen by comparing FID, LPIPS, SSIM, and PSNR over values 1.0, 1.1, 1.25, and 1.4 on the test set (Tables 4 and 5), then used for the final reported results.
  • NFE (number of function evaluations) = 60 for medium speed test, 55 at 512x512, 30 or 60 for large; ablations use 60 and 720
    Selected by hand to trade speed against quality; the paper reports multiple NFE values and chooses specific ones for the headline speed and quality numbers.
assumptions (5)
  • standard math Flow matching with the conditional optimal transport path is a valid generative objective (Eq. 3-4).
    Invoked from Lipman et al. [15] in Section 3.1; the paper uses the standard objective without modification.
  • domain assumption Pretrained VAE and DINOv2 encoders preserve the information needed for appearance and pose conditioning.
    Section 3.2 relies on these encoders to transfer source appearance and target pose into the latent and embedding space; no person-specific fine-tuning of the encoders is described.
  • domain assumption DeepFashion with OpenPose 18-keypoint RGB skeleton poses is a representative benchmark for PGPIS.
    Section 4.1 defines the evaluation; the paper draws conclusions about real-time applications from this single dataset.
  • domain assumption Batch-of-8 runtime on one A100 under the X-MDPT protocol is representative of real-time video generation.
    Section 4.3 extrapolates from batch latency to frame-by-frame video generation; this assumes batching pose frames is a valid deployment mode.
  • domain assumption The DiT backbone is faster than U-Net for this task.
    Section 3.2 cites LFM for this claim and excludes U-Net baselines such as PoCoLD from the speed comparison; it is not independently verified in this paper.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Real-Time Person Image Synthesis Using a Flow Matching Model." pith.science (2026). https://pith.science/paper/IXQ52YFD

@misc{pith2026250503562,
  author       = {Pith},
  title        = {Pith review of: Real-Time Person Image Synthesis Using a Flow Matching Model},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/IXQ52YFD}},
  note         = {Machine review of arXiv:2505.03562}
}
read the original abstract

Pose-Guided Person Image Synthesis (PGPIS) generates realistic person images conditioned on a target pose and a source image. This task plays a key role in various real-world applications, such as sign language video generation, AR/VR, gaming, and live streaming. In these scenarios, real-time PGPIS is critical for providing immediate visual feedback and maintaining user immersion.However, achieving real-time performance remains a significant challenge due to the complexity of synthesizing high-fidelity images from diverse and dynamic human poses. Recent diffusion-based methods have shown impressive image quality in PGPIS, but their slow sampling speeds hinder deployment in time-sensitive applications. This latency is particularly problematic in tasks like generating sign language videos during live broadcasts, where rapid image updates are required. Therefore, developing a fast and reliable PGPIS model is a crucial step toward enabling real-time interactive systems. To address this challenge, we propose a generative model based on flow matching (FM). Our approach enables faster, more stable, and more efficient training and sampling. Furthermore, the proposed model supports conditional generation and can operate in latent space, making it especially suitable for real-time PGPIS applications where both speed and quality are critical. We evaluate our proposed method, Real-Time Person Image Synthesis Using a Flow Matching Model (RPFM), on the widely used DeepFashion dataset for PGPIS tasks. Our results show that RPFM achieves near-real-time sampling speeds while maintaining performance comparable to the state-of-the-art models. Our methodology trades off a slight, acceptable decrease in generated-image accuracy for over a twofold increase in generation speed, thereby ensuring real-time performance.

Figures

Figures reproduced from arXiv: 2505.03562 by the authors.

Figure 1
Figure 1. An illustration of the DeepFashion PGPIS dataset is [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. The architecture of our Real-Time Person Image Synthesis employs a Flow Matching Model. This architecture encompasses [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. visualization of Tab [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Our model produces images that correspond to the in [PITH_FULL_IMAGE:figures/full_fig_p009_4.png]
Figure 5
Figure 5. Figure 5: A qualitative comparative study with baselines (in the DeepFashion dataset) [PITH_FULL_IMAGE:figures/full_fig_p013_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

35 extracted references · 23 canonical work pages

  1. [1]

    Person image synthesis via de- noising diffusion model

    Ankan Kumar Bhunia, Salman Khan, Hisham Cholakkal, Rao Muhammad Anwer, Jorma Laaksonen, Mubarak Shah, and Fahad Shahbaz Khan. Person image synthesis via de- noising diffusion model. In CVPR, pages 5968–5976, 2023. 1, 3, 4, 6, 7

    work page 2023

  2. [2]

    Neural ordinary differential equa- tions

    Ricky TQ Chen, Yulia Rubanova, Jesse Bettencourt, and David K Duvenaud. Neural ordinary differential equa- tions. Advances in neural information processing systems , 31, 2018. 2, 3

    work page 2018

  3. [3]

    Flow matching in latent space

    Quan Dao, Hao Phung, Binh Nguyen, and Anh Tran. Flow matching in latent space. arXiv preprint arXiv:2307.08698,

    arXiv

  4. [4]

    Taming transformers for high-resolution image synthesis

    Patrick Esser, Robin Rombach, and Bjorn Ommer. Taming transformers for high-resolution image synthesis. In Pro- ceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 12873–12883, 2021. 4

    2021

  5. [5]

    Generative adversarial networks

    Ian Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courville, and Yoshua Bengio. Generative adversarial networks. Commu- nications of the ACM, 63(11):139–144, 2020. 4

    2020

  6. [6]

    Controllable person image synthesis with pose-constrained latent diffusion

    Xiao Han, Xiatian Zhu, Jiankang Deng, Yi-Zhe Song, and Tao Xiang. Controllable person image synthesis with pose-constrained latent diffusion. In Proceedings of the IEEE/CVF International Conference on Computer Vision , pages 22768–22777, 2023. 6

    work page 2023

  7. [7]

    Deep residual learning for image recognition

    Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Deep residual learning for image recognition. In Proceed- ings of the IEEE conference on computer vision and pattern recognition, pages 770–778, 2016. 2

    2016

  8. [8]

    Gans trained by a two time-scale update rule converge to a local nash equilib- rium

    Martin Heusel, Hubert Ramsauer, Thomas Unterthiner, Bernhard Nessler, and Sepp Hochreiter. Gans trained by a two time-scale update rule converge to a local nash equilib- rium. Advances in neural information processing systems , 30, 2017. 6

    2017

Show all 35 references
  1. [9]

    Classifier-free diffusion guidance

    Jonathan Ho and Tim Salimans. Classifier-free diffusion guidance. In NeurIPS Workshop, 2021. 6

    2021

  2. [10]

    Denoising dif- fusion probabilistic models

    Jonathan Ho, Ajay Jain, and Pieter Abbeel. Denoising dif- fusion probabilistic models. Advances in neural information processing systems, 33:6840–6851, 2020. 2, 3

    2020

  3. [11]

    Motion flow matching for human motion synthesis and editing

    Vincent Tao Hu, Wenzhe Yin, Pingchuan Ma, Yunlu Chen, Basura Fernando, Yuki M Asano, Efstratios Gavves, Pascal Mettes, Bjorn Ommer, and Cees GM Snoek. Motion flow matching for human motion synthesis and editing. arXiv preprint arXiv:2312.08895, 2023. 2

    2023 arXiv

  4. [12]

    A survey on gans for computer vision: Recent re- search, analysis and taxonomy

    Guillermo Iglesias, Edgar Talavera, and Alberto D ´ıaz- ´Alvarez. A survey on gans for computer vision: Recent re- search, analysis and taxonomy. Computer Science Review , 48:100553, 2023. 4

    2023

  5. [13]

    V oicebox: Text-guided multilingual universal speech generation at scale

    Matthew Le, Apoorv Vyas, Bowen Shi, Brian Karrer, Leda Sari, Rashel Moritz, Mary Williamson, Vimal Manohar, Yossi Adi, Jay Mahadeokar, et al. V oicebox: Text-guided multilingual universal speech generation at scale. Advances in neural information processing systems, 36, 2024. 2, 4

    2024

  6. [14]

    Appear- ance and pose-guided human generation: A survey

    Fangjian Liao, Xingxing Zou, and Waikeung Wong. Appear- ance and pose-guided human generation: A survey. ACM Computing Surveys, 56(5):1–35, 2024. 1

    2024

  7. [15]

    Flow matching for generative mod- eling

    Yaron Lipman, Ricky TQ Chen, Heli Ben-Hamu, Maximil- ian Nickel, and Matt Le. Flow matching for generative mod- eling. In ICLR, 2023. 2, 3, 4

    2023

  8. [16]

    Generative pre-training for speech with flow matching

    Alexander H Liu, Matt Le, Apoorv Vyas, Bowen Shi, An- dros Tjandra, and Wei-Ning Hsu. Generative pre-training for speech with flow matching. In ICLR, 2024. 2, 4

    2024

  9. [17]

    Deepfashion: Powering robust clothes recognition and retrieval with rich annotations

    Ziwei Liu, Ping Luo, Shi Qiu, Xiaogang Wang, and Xiaoou Tang. Deepfashion: Powering robust clothes recognition and retrieval with rich annotations. In Proceedings of the IEEE conference on computer vision and pattern recogni- tion, pages 1096–1104, 2016. 6

    2016

  10. [18]

    Coarse-to-fine latent diffusion for pose- guided person image synthesis

    Yanzuo Lu, Manlin Zhang, Andy J Ma, Xiaohua Xie, and Jianhuang Lai. Coarse-to-fine latent diffusion for pose- guided person image synthesis. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 6420–6429, 2024. 1, 3, 4, 6, 7

    2024

  11. [19]

    Learning semantic per- son image generation by region-adaptive normalization

    Zhengyao Lv, Xiaoming Li, Xin Li, Fu Li, Tianwei Lin, Dongliang He, and Wangmeng Zuo. Learning semantic per- son image generation by region-adaptive normalization. In Proceedings of the IEEE/CVF Conference on Computer Vi- sion and Pattern Recognition, pages 10806–10815, 2021. 6, 1

    2021

  12. [20]

    Pose guided person image gener- ation

    Liqian Ma, Xu Jia, Qianru Sun, Bernt Schiele, Tinne Tuyte- laars, and Luc Van Gool. Pose guided person image gener- ation. Advances in neural information processing systems , 30, 2017. 3

    2017

  13. [21]

    Dinov2: Learning robust visual features without supervision

    Maxime Oquab, Timoth ´ee Darcet, Th ´eo Moutakanni, Huy V o, Marc Szafraniec, Vasil Khalidov, Pierre Fernandez, Daniel Haziza, Francisco Massa, Alaaeldin El-Nouby, et al. Dinov2: Learning robust visual features without supervision. In TMLR, pages 12873–12883. 5

  14. [22]

    Scalable diffusion models with transformers

    William Peebles and Saining Xie. Scalable diffusion models with transformers. In Proceedings of the IEEE/CVF Inter- national Conference on Computer Vision, pages 4195–4205,

  15. [23]

    Cross-view masked diffusion transformers for person image synthesis

    Trung X Pham, Zhang Kang, and Chang D Yoo. Cross-view masked diffusion transformers for person image synthesis

  16. [24]

    Neural texture extraction and distribution for controllable person image synthesis

    Yurui Ren, Xiaoqing Fan, Ge Li, Shan Liu, and Thomas H Li. Neural texture extraction and distribution for controllable person image synthesis. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages 13535–13544, 2022. 4, 6, 1

    2022

  17. [25]

    Variational inference with normalizing flows

    Danilo Rezende and Shakir Mohamed. Variational inference with normalizing flows. In International conference on ma- chine learning, pages 1530–1538. PMLR, 2015. 3

    2015

  18. [26]

    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 Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 10684–10695, 2022. 4

    2022

  19. [27]

    U- net: Convolutional networks for biomedical image segmen- tation

    Olaf Ronneberger, Philipp Fischer, and Thomas Brox. U- net: Convolutional networks for biomedical image segmen- tation. In Medical image computing and computer-assisted intervention–MICCAI 2015: 18th international conference, Munich, Germany, October 5-9, 2015, proceedings, pa...

    2015

  20. [28]

    Deep person generation: A survey from the perspective of face, pose, and cloth synthesis

    Tong Sha, Wei Zhang, Tong Shen, Zhoujun Li, and Tao Mei. Deep person generation: A survey from the perspective of face, pose, and cloth synthesis. ACM Computing Surveys, 55 (12):1–37, 2023. 1

    2023

  21. [29]

    Advancing pose-guided image synthesis with progres- sive conditional diffusion models

    Fei Shen, Hu Ye, Jun Zhang, Cong Wang, Xiao Han, and Wei Yang. Advancing pose-guided image synthesis with progres- sive conditional diffusion models. In ICLR, 2024. 1, 3

    2024

  22. [30]

    Image quality assessment: from error visibility to structural similarity

    Zhou Wang, Alan C Bovik, Hamid R Sheikh, and Eero P Si- moncelli. Image quality assessment: from error visibility to structural similarity. IEEE transactions on image processing, 13(4):600–612, 2004. 6

    2004

  23. [31]

    Exploring dual-task correlation for pose guided person image generation

    Pengze Zhang, Lingxiao Yang, Jian-Huang Lai, and Xiaohua Xie. Exploring dual-task correlation for pose guided person image generation. In Proceedings of the IEEE/CVF Con- ference on Computer Vision and Pattern Recognition, pages 7713–7722, 2022. 4, 6, 1

    2022

  24. [32]

    The unreasonable effectiveness of deep features as a perceptual metric

    Richard Zhang, Phillip Isola, Alexei A Efros, Eli Shecht- man, and Oliver Wang. The unreasonable effectiveness of deep features as a perceptual metric. In Proceedings of the IEEE conference on computer vision and pattern recogni- tion, pages 586–595, 2018. 6

    2018

  25. [33]

    Cocosnet v2: Full-resolution correspondence learning for image trans- lation

    Xingran Zhou, Bo Zhang, Ting Zhang, Pan Zhang, Jianmin Bao, Dong Chen, Zhongfei Zhang, and Fang Wen. Cocosnet v2: Full-resolution correspondence learning for image trans- lation. In Proceedings of the IEEE/CVF conference on com- puter vision and pattern recognition , pages 114...

  26. [34]

    Cross attention based style distribution for controllable person image synthesis

    Xinyue Zhou, Mingyu Yin, Xinyuan Chen, Li Sun, Changxin Gao, and Qingli Li. Cross attention based style distribution for controllable person image synthesis. In European Con- ference on Computer Vision, pages 161–178. Springer, 2022. 4, 6, 1

    2022

  27. [35]

    Progressive pose attention transfer for person image generation

    Zhen Zhu, Tengteng Huang, Baoguang Shi, Miao Yu, Bofei Wang, and Xiang Bai. Progressive pose attention transfer for person image generation. In CVPR, pages 2347–2356, 2019. 1 Real-Time Person Image Synthesis Using a Flow Matching Model Supplementary Material Qualitative compar...

    2019

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.