Pith. sign in

REVIEW 2 major objections 6 minor 46 references

Fusion Embedding for Pose-Guided Person Image Synthesis with Diffusion Model

T0 review · 2 major / 6 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read This paper proposes FPDM, a two-stage diffusion model that explicitly aligns a fused source-pose embedding with the target image embedding via contrastive learning, then uses that embedding to condition image generation.

desk verdict Solid two-stage PGPIS pipeline with a real but incremental contrastive fusion idea; the headline ablation claim is unsupported by tiny, unreplicated deltas. read the letter →

arxiv 2412.07333 v2 pith:HKE6NBRC submitted 2024-12-10 cs.CV cs.AI

Donghwna Lee , Kirok Kim , Jisu Lee , Kyungha Min , Wooju Kim This is my paper · ORCID
classification cs.CVcs.AI
keywords pose-guidedpersonimagesynthesisdiffusionmodelcontrastivelearningfusionembeddingsource-enhancedposelatentsignlanguagegenerationDeepFashion
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

FPDM is a two-stage diffusion method for pose-guided person image synthesis whose central claim is that generation is best driven by an explicit fusion embedding of the source image and the target pose, learned to match the target image's embedding before the diffusion model sees it. The first stage trains an image-pose fusion module with a contrastive loss that also pulls the source embedding into the batch, so the learned embedding separates person identity from pose. The second stage feeds the frozen fusion embedding into a latent diffusion U-Net together with source-image features and pose features, and the paper reports state-of-the-art LPIPS, SSIM, and PSNR on DeepFashion plus large gains over prior sign-language generation methods on the PHOENIX-2014T benchmark. The claimed payoff is consistency: the same target pose yields stable imagery across different source images, and the same source yields stable style across different poses.

What carries the argument

The Image-Pose Fusion (IPF) module is the load-bearing mechanism: it uses a CLIP vision encoder to embed source image, target pose, and target image, combines source and pose embeddings with a Combiner module, and trains the result against the target embedding with InfoNCE contrastive loss. The paper's Source-Enhanced Pose Fusion variant doubles the batch by treating each source embedding as its own target, which forces the embedding space to separate persons by pose rather than by identity. The second stage then conditions a latent diffusion U-Net with three inputs: DINOv2 features of the source image acting as keys and values in transformer blocks, a four-layer convolutional pose encoder following the ControlNet design, and the fusion embedding added to the timestep embedding in ResNet blocks. Classifier-free guidance is applied with separate weights for the source-pose conditioning and the fusion embedding.

What would settle it

Train the same FPDM pipeline twice, once with the three-channel RGB pose image and once with a multi-channel pose encoding that separates joints or joint connections, keeping all other settings identical; if the RGB variant does not match or beat the multi-channel variant on pose-alignment metrics such as hand-keypoint error or LPIPS on complex poses, the paper's central conditioning premise fails. A second check is retrieval accuracy of the first stage on a held-out pose set: the R@1 claim of 0.997 should be reproduced when pose images are drawn from unseen signers or unusual body configurations.

Watch

Extended reading notes

Core claim

The paper claims to be the first to explicitly align a fused source-pose embedding with the target image embedding via contrastive learning and then use that aligned embedding as a conditioning signal for pose-guided diffusion. On DeepFashion, its best model reaches LPIPS 0.1445, SSIM 0.7417, and PSNR 18.832 at 256x176, outperforming the compared state of the art on those semantic and pixel-level metrics, and on the PHOENIX sign-language set it reports SSIM 0.886 and FID 5.129 against 0.759 and 27.75 for the best prior method. The first-stage retrieval experiment, with R@1 of 0.997, is presented as evidence that the fusion embedding is genuinely aligned with the target. The paper also claims robustness: images stay consistent when the source image or the target pose varies.

Load-bearing premise

The load-bearing premise is that a single three-channel RGB skeleton image of the target pose carries enough geometric detail for both the CLIP encoder and the four-layer CNN pose encoder to drive accurate pose-controlled generation, even though the paper does not quantitatively compare this representation against multi-channel per-joint pose encodings.

Editorial extensions

If this is right

  • The two-stage recipe yields state-of-the-art LPIPS, SSIM, and PSNR on DeepFashion at 256x176 and best LPIPS at 512x352, so texture fidelity and pixel alignment improve without waiting for implicit diffusion learning.
  • The same model transfers to sign-language frames, reporting SSIM 0.886 and FID 5.129 on PHOENIX-2014T, far ahead of prior video and generation baselines, which suggests the conditioning mechanism works where precise hand poses matter.
  • The stage-two-only model already approaches state-of-the-art performance, so the paper offers its structure as a simple, reusable baseline for future pose-guided person image synthesis diffusion models.
  • Because the fusion embedding is aligned to the target before generation, the model is robust to source-image and pose variations, as shown in the qualitative ablations.
  • The first-stage fusion retrieval reaches R@1 of 0.997, meaning the embedding alone nearly identifies the correct target image, directly supporting its use as a generation condition.

Reading between the lines

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

  • Editorial inference: the contrastively aligned fusion embedding is a general conditioning idea, not tied to person images; the same two-stage alignment could be tested for object reposing, virtual try-on with different garment views, or any image geometry-to-appearance task.
  • Editorial inference: the paper's own tables show the B3-to-B6 differences are modest, so the headline gains on DeepFashion may owe more to scaling the source encoder input to 512 than to the fusion embedding itself; ablations that hold the source encoder fixed while toggling only the IPF loss would isolate this.
  • Editorial inference: the shared CLIP encoder for images and poses is a resource-driven choice; separate encoders fine-tuned for pose geometry would make the RGB-pose assumption more testable and could improve pose fidelity on extreme angles.
  • Editorial inference: a testable extension is using the learned fusion embedding for zero-shot or few-shot pose transfer across unseen identities, since the source-enhanced contrastive objective trains the embedding to be identity-independent.
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

2 major / 6 minor

Summary. The paper proposes FPDM, a two-stage framework for pose-guided person image synthesis. In the first stage, CLIP image encoders and a Combiner module fuse the source image embedding and a target pose image embedding into a single fusion embedding, trained with a source-enhanced InfoNCE contrastive loss to align with the target image embedding. In the second stage, a Stable Diffusion backbone is conditioned on DINOv2 source-image features, a four-layer CNN pose encoding, and the frozen fusion embedding, with classifier-free guidance applied jointly over appearance, pose, and fusion conditions. The method is evaluated on DeepFashion and RWTH-PHOENIX-Weather 2014T, reporting state-of-the-art LPIPS, SSIM, and PSNR on DeepFashion and large improvements on Phoenix, along with ablations of the fusion module and pose/source encoders.

Significance. If the reported results are reproducible, the paper offers a practically useful conditioning mechanism for diffusion-based pose-guided person image synthesis. The authors deserve credit for releasing code, for introducing FIDt to address the known DeepFashion train/test distribution shift, and for reporting a first-stage retrieval result (R@1 = 0.997) that independently demonstrates embedding-space alignment. The Phoenix sign-language evaluation is also a valuable extension. However, the central causal claim that explicit fusion embedding alignment 'substantially improves' generation is not supported by the second-stage ablation, whose deltas are small, unreplicated, and partly reverse in FID; the Phoenix hand-metric comparison is weakened by a detector mismatch; and the assertion that a three-channel RGB pose image suffices is only visually verified, with no quantitative pose-representation ablation. These are load-bearing issues for the manuscript's stated novelty claims.

major comments (2)
  1. [Quantitative Results for Sign Language Image Generation (Table 4)] The Hand PE comparison in Table 4 is not apples-to-apples: SignGAN's Hand PE is computed with the hand pose detector of Ge et al., while FPDM uses HRNetv2. Hand PE is a keypoint-distance metric and is known to be sensitive to the detector used to extract the keypoints. The authors should recompute at least one of the two methods with the other detector and report both values, or otherwise state clearly that the Hand PE numbers are not directly comparable. This matters because the Phoenix table is one of the two main empirical pillars of the paper.
  2. [Fusion Embedding Conditioned Diffusion Model (Method)] The claim that the first-stage retrieval result (A3, R@1 = 0.997) establishes a generation benefit is missing a link. Retrieval accuracy shows that the fusion embedding is well aligned in the CLIP embedding space, but it does not by itself prove that injecting that embedding into the second-stage diffusion model improves synthesis. The second-stage ablation is the relevant test, and as noted above its gains are small and unreplicated. The authors should either provide a stronger second-stage ablation or explicitly frame the retrieval result as evidence of alignment rather than as evidence of generation quality.
minor comments (6)
  1. [Method] The sentence 'In the first stage In the first stage, the CLIP image encoder is used...' contains a duplicated phrase that should be removed.
  2. [Related Work] The paragraph describing PCDM, RePoseDM, and X-MDPT repeats the same sentence twice ('PCDM ... use Transformer to perform PGPIS by deeply learning the relationship between the source and the target pose images').
  3. [Ablation Study] Table 2 is captioned 'Qualitative results of first-stage ablation' but reports quantitative Recall@k and average rank; the caption should say 'Quantitative results.'
  4. [Preliminary (Equation 1)] Equation (1) writes 'ϵ N(0, 1)', missing the distribution symbol; it should be 'ϵ ~ N(0, 1)'.
  5. [Experiments (Metrics)] The definition of FIDt is given only as 'measures the FID of the test dataset'; please specify precisely which feature distributions are compared and how the train/test shift is handled, so that the metric can be reproduced by other groups.
  6. [Figure 1] The figure is referenced inconsistently as 'Figure 1(a)' and 'Figure 1 (B)' in the text; the labels should be unified.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: FPDM's fusion embedding is trained and evaluated on held-out benchmarks; the reported generation results are empirical, not derived from their inputs.

full rationale

FPDM is an empirical two-stage pipeline: the Image-Pose Fusion (IPF) module is trained with an InfoNCE contrastive loss to align fused source+pose embeddings with target image embeddings, and the diffusion model is trained with an MSE denoising objective conditioned on source, pose, and fusion embeddings. At inference, the fusion embedding is computed from the source image and target pose only; the target image is used solely as supervision, never as a generation input. The reported numbers are held-out test evaluations on DeepFashion and RWTH-PHOENIX-Weather 2014T against external baselines, so the central claim is not a renamed input or a fitted constant recycled as a prediction. The ablation study (Table 3) compares internally consistent variants; B6 versus B4/B5 shows small deltas, and the Limitations section itself concedes the differences are 'noticeable but not substantial.' That concession weakens the abstract's wording 'substantially improves,' but it is a claim-strength/statistical-evidence issue, not circularity. The first-stage retrieval result A3 R@1 = 0.997 is a direct measurement of the trained contrastive objective, not a derived prediction used to fabricate a generation result. No load-bearing self-citation appears: cited prior works (Baldrati, Shen, Rombach, etc.) are external, and no uniqueness theorem or ansatz is imported from the present authors' prior work. Guidance weights and hyperparameters are standard sampling/training choices, not fitted to the test metrics. Overall, the derivation chain is self-contained and the paper's claims are falsifiable against held-out data.

Assumptions & free parameters 2 free parameters · 3 assumptions · 0 invented entities

The paper introduces no new physical entities. The central claim rests on two domain assumptions about representation sufficiency (RGB pose encoding and CLIP embedding space) and one about fine-tuning stability. The only hand-chosen free parameters are standard guidance and training hyperparameters. No parameter is fitted to the test set to produce the reported numbers.

free parameters (2)
  • Guidance weights w_c and w_f = 2.0 and 2.0
    Set during sampling and used in Eq. 8 for classifier-free guidance. These are hand-chosen and directly affect generation quality; no tuning procedure is reported.
  • Condition-drop probability = 20%
    Probability of dropping e'_s, e'_tp, and ef during training for classifier-free guidance. A standard hyperparameter, but chosen by hand.
assumptions (3)
  • domain assumption A target pose rendered as a single three-channel RGB skeleton image contains sufficient geometric information for both the CLIP encoder and the CNN pose encoder to control generation.
    Invoked in the Method section where the authors state they 'visually verified' the effectiveness of 3-channel RGB pose images. No quantitative comparison against multi-channel pose encodings is provided.
  • domain assumption CLIP image embeddings provide a shared semantic space where a fused source-pose embedding can be aligned to a target image embedding via contrastive learning.
    The entire first stage depends on the premise that CLIP features of source, pose, and target images are commensurable. This is assumed and validated only through retrieval accuracy on the training distribution.
  • domain assumption Fine-tuning Stable Diffusion 2.1 with an additional fusion embedding condition preserves the generative prior of the pre-trained model.
    The second stage assumes that injecting the learned fusion embedding into the UNet's ResNet blocks and using source features as key/value in transformer blocks does not degrade the pre-trained Stable Diffusion capabilities. This is an empirical premise, not proven by analysis.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Fusion Embedding for Pose-Guided Person Image Synthesis with Diffusion Model." pith.science (2026). https://pith.science/paper/HKE6NBRC

@misc{pith2026241207333,
  author       = {Pith},
  title        = {Pith review of: Fusion Embedding for Pose-Guided Person Image Synthesis with Diffusion Model},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/HKE6NBRC}},
  note         = {Machine review of arXiv:2412.07333}
}
read the original abstract

Pose-Guided Person Image Synthesis (PGPIS) aims to generate human images in specified poses while preserving the identity and appearance of a source image. This technology facilitates diverse applications, including virtual try-on, digital avatars, animation, and sign language generation. Despite the high-quality results of recent diffusion-based PGPIS, these models typically depend on implicit feature aggregation within the denoising process. As a result, fine-grained texture preservation is limited, and even for the same identity, it is difficult to ensure consistent generation under variations in pose and source appearance. To address these limitations, we propose Fusion Embedding for PGPIS using a Diffusion Model (FPDM), the first framework that explicitly aligns fused source-pose embeddings with target image embeddings via contrastive learning, and subsequently employs the learned fusion embedding as a conditioning signal for generation. FPDM integrates an Image-Pose Fusion (IPF) module into our proposed Source-Enhanced Pose Fusion approach to learn a fusion embedding aligned with the target image. We then employ a conditional diffusion model guided by source appearance, target pose, and the learned fusion embedding. Experiments on the DeepFashion benchmark and the RWTH-PHOENIX-Weather 2014T dataset demonstrate competitive performance compared to existing methods in both quantitative and qualitative evaluations, with ablation studies confirming that explicit fusion embedding alignment substantially improves texture fidelity and consistency across pose and source appearance variations.

Figures

Figures reproduced from arXiv: 2412.07333 by the authors.

Figure 1
Figure 1. Architecture of the Fusion Embedding for PGPIS with Diffusion Model. [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 3
Figure 3. Visualization of the robustness of generated im [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figure 2
Figure 2. Qualitative comparisons with current state-of-the [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figures from the paper (2 more)
Figure 5
Figure 5. Figure 5: Qualitative comparisons of second-stage ablation [PITH_FULL_IMAGE:figures/full_fig_p007_5.png]
Figure 6
Figure 6. Figure 6: Effect of Fusion Embedding in Ablation Studies [PITH_FULL_IMAGE:figures/full_fig_p007_6.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

46 extracted references · 33 canonical work pages

  1. [1]

    , " * write output.state after.block = add.period write newline

    ENTRY address archivePrefix author booktitle chapter edition editor eid eprint howpublished institution isbn journal key month note number organization pages publisher school series title type volume year label extra.label sort.label short.list INTEGERS output.state before.all mid.sentence after.sentence after.block FUNCTION init.state.consts #0 'before.a...

  2. [2]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 global.max substring 't := if while FUNCTION word.in bbl.in capitalize " " * FUNCT...

  3. [3]

    Albahar, B.; Lu, J.; Yang, J.; Shu, Z.; Shechtman, E.; and Huang, J.-B. 2021. Pose with style: detail-preserving pose-guided image synthesis with conditional StyleGAN. ACM Trans. Graph., 40(6): Article 218

    work page 2021

  4. [4]

    Baldrati, A.; Bertini, M.; Uricchio, T.; and Bimbo, A. D. 2023. Composed Image Retrieval using Contrastive Learning and Task-oriented CLIP-based Features. ACM Trans. Multimedia Comput. Commun. Appl., 20(3): Article 62

    work page 2023

  5. [5]

    K.; Khan, S.; Cholakkal, H.; Anwer, R

    Bhunia, A. K.; Khan, S.; Cholakkal, H.; Anwer, R. M.; Laaksonen, J.; Shah, M.; and Khan, F. S. 2023. Person Image Synthesis via Denoising Diffusion Model. In IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 5968--5976. ISBN 2575-7075

    work page 2023

  6. [6]

    C.; Hadfield, S.; Koller, O.; Ney, H.; and Bowden, R

    Camgoz, N. C.; Hadfield, S.; Koller, O.; Ney, H.; and Bowden, R. 2018. Neural Sign Language Translation. In IEEE/CVF Conference on Computer Vision and Pattern Recognition, 7784--7793. ISBN 2575-7075

    work page 2018

  7. [7]

    Chan, C.; Ginosar, S.; Zhou, T.; and Efros, A. 2019. Everybody Dance Now. In IEEE/CVF International Conference on Computer Vision (ICCV), 5932--5941. ISBN 2380-7504

    work page 2019

  8. [8]

    Dhariwal, P.; and Nichol, A. 2021. Diffusion models beat gans on image synthesis. Advances in neural information processing systems, 34: 8780--8794

    2021

Show all 46 references
  1. [9]

    Dosovitskiy, A.; Beyer, L.; Kolesnikov, A.; Weissenborn, D.; Zhai, X.; Unterthiner, T.; Dehghani, M.; Minderer, M.; Heigold, G.; and Gelly, S. 2020. An image is worth 16x16 words: Transformers for image recognition at scale. arXiv preprint arXiv:2010.11929

    2020 arXiv

  2. [10]

    Esser, P.; Rombach, R.; and Ommer, B. 2020. Taming transformers for high-resolution image synthesis. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 12873--12883

    2020

  3. [11]

    Ge, L.; Ren, Z.; Li, Y.; Xue, Z.; Wang, Y.; Cai, J.; and Yuan, J. 2019. 3D Hand Shape and Pose Estimation From a Single RGB Image. IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 10825--10834

    2019

  4. [12]

    Goodfellow, I.; Pouget-Abadie, J.; Mirza, M.; Xu, B.; Warde-Farley, D.; Ozair, S.; Courville, A.; and Bengio, Y. 2020. Generative adversarial networks. Commun. ACM, 63(11): 139–144

    2020

  5. [13]

    Guan, M.; Wang, Y.; Ma, G.; Liu, J.; and Sun, M. 2024. Multi-Stream Keypoint Attention Network for Sign Language Recognition and Translation. arXiv preprint arXiv:2405.05672

    2024 arXiv

  6. [14]

    Han, X.; Zhu, X.; Deng, J.; Song, Y.-Z.; and Xiang, T. 2023. Controllable person image synthesis with pose-constrained latent diffusion. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 22768--22777

    2023

  7. [15]

    Heusel, M.; Ramsauer, H.; Unterthiner, T.; Nessler, B.; and Hochreiter, S. 2017. GANs trained by a two time-scale update rule converge to a local nash equilibrium

    2017

  8. [16]

    Ho, J.; Jain, A.; and Abbeel, P. 2020. Denoising diffusion probabilistic models. Advances in neural information processing systems, 33: 6840--6851

    2020

  9. [17]

    Ho, J.; and Salimans, T. 2021. Classifier-Free Diffusion Guidance

    2021

  10. [18]

    Iglesias, G.; Talavera, E.; and Díaz-Álvarez, A. 2023. A survey on GANs for computer vision: Recent research, analysis and taxonomy. Computer Science Review, 48: 100553

    2023

  11. [19]

    Khandelwal, A. 2024. RePoseDM: Recurrent Pose Alignment and Gradient Guidance for Pose Guided Image Synthesis. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2495--2504

    2024

  12. [20]

    Kingma, D.; Salimans, T.; Poole, B.; and Ho, J. 2021. Variational diffusion models. Advances in neural information processing systems, 34: 21696--21707

    2021

  13. [21]

    P.; and Welling, M

    Kingma, D. P.; and Welling, M. 2013. Auto-encoding variational bayes. arXiv preprint arXiv:1312.6114

    2013 arXiv

  14. [22]

    Liao, F.; Zou, X.; and Wong, W. 2024. Appearance and Pose-guided Human Generation: A Survey. ACM Comput. Surv., 56(5): Article 129

    2024

  15. [23]

    Liu, Z.; Luo, P.; Qiu, S.; Wang, X.; and Tang, X. 2016. DeepFashion: Powering Robust Clothes Recognition and Retrieval with Rich Annotations. In IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 1096--1104. ISBN 1063-6919

    2016

  16. [24]

    J.; Xie, X.; and Lai, J.-H

    Lu, Y.; Zhang, M.; Ma, A. J.; Xie, X.; and Lai, J.-H. 2024. Coarse-to-Fine Latent Diffusion for Pose-Guided Person Image Synthesis. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), volume abs/2402.18078

    2024 arXiv

  17. [25]

    Lv, Z.; Li, X.; Li, X.; Li, F.; Lin, T.; He, D.; and Zuo, W. 2021. Learning semantic person image generation by region-adaptive normalization. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 10806--10815

    2021

  18. [26]

    Ma, L.; Jia, X.; Sun, Q.; Schiele, B.; Tuytelaars, T.; and Van Gool, L. 2017. Pose guided person image generation. Advances in neural information processing systems, 30

    2017

  19. [27]

    Oord, A. v. d.; Li, Y.; and Vinyals, O. 2018. Representation learning with contrastive predictive coding. arXiv preprint arXiv:1807.03748

    2018 arXiv

  20. [28]

    Paszke, A.; Gross, S.; Massa, F.; Lerer, A.; Bradbury, J.; Chanan, G.; Killeen, T.; Lin, Z.; Gimelshein, N.; and Antiga, L. 2019. Pytorch: An imperative style, high-performance deep learning library. Advances in neural information processing systems, 32

    2019

  21. [29]

    X.; Zhang, K.; and Yoo, C

    Pham, T. X.; Zhang, K.; and Yoo, C. D. 2024. Cross-view Masked Diffusion Transformers for Person Image Synthesis

    2024

  22. [30]

    W.; Hallacy, C.; Ramesh, A.; Goh, G.; Agarwal, S.; Sastry, G.; Askell, A.; Mishkin, P.; and Clark, J

    Radford, A.; Kim, J. W.; Hallacy, C.; Ramesh, A.; Goh, G.; Agarwal, S.; Sastry, G.; Askell, A.; Mishkin, P.; and Clark, J. 2021. Learning transferable visual models from natural language supervision. In International conference on machine learning, 8748--8763. PMLR. ISBN 2640-3498

    2021

  23. [31]

    Ren, Y.; Fan, X.; Li, G.; Liu, S.; and Li, T. H. 2022. Neural Texture Extraction and Distribution for Controllable Person Image Synthesis. In IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 13525--13534. ISBN 2575-7075

    2022

  24. [32]

    Rombach, R.; Blattmann, A.; Lorenz, D.; Esser, P.; and Ommer, B. 2021. High-resolution image synthesis with latent diffusion models. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 10684--10695

    2021

  25. [33]

    Ronneberger, O.; Fischer, P.; and Brox, T. 2015. U-Net: Convolutional Networks for Biomedical Image Segmentation. In Navab, N.; Hornegger, J.; Wells, W. M.; and Frangi, A. F., eds., Medical Image Computing and Computer-Assisted Intervention, 234--241. Springer International Pu...

    2015

  26. [34]

    C.; and Bowden, R

    Saunders, B.; Camgoz, N. C.; and Bowden, R. 2022. Signing at scale: Learning to co-articulate signs for large-scale photo-realistic sign language production. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 5141--5151

    2022

  27. [35]

    Sha, T.; Zhang, W.; Shen, T.; Li, Z.; and Mei, T. 2023. Deep Person Generation: A Survey from the Perspective of Face, Pose, and Cloth Synthesis. ACM Comput. Surv., 55(12): Article 257

    2023

  28. [36]

    Shen, F.; Ye, H.; Zhang, J.; Wang, C.; Han, X.; and Yang, W. 2024. Advancing pose-guided image synthesis with progressive conditional diffusion models. In The Twelfth International Conference on Learning Representations

    2024

  29. [37]

    Stoll, S.; Camgoz, N.; Hadfield, S.; and Bowden, R. 2020. Text2Sign: Towards Sign Language Production Using Neural Machine Translation and Generative Adversarial Networks. International Journal of Computer Vision

    2020

  30. [38]

    Wang, J.; Sun, K.; Cheng, T.; Jiang, B.; Deng, C.; Zhao, Y.; Liu, D.; Mu, Y.; Tan, M.; and Wang, X. 2020. Deep high-resolution representation learning for visual recognition. IEEE transactions on pattern analysis and machine intelligence, 43(10): 3349--3364

    2020

  31. [39]

    Wang, T.-C.; Liu, M.-Y.; Zhu, J.-Y.; Liu, G.; Tao, A.; Kautz, J.; and Catanzaro, B. 2018 a . Video-to-video synthesis. arXiv preprint arXiv:1808.06601

    2018 arXiv

  32. [40]

    C.; Liu, M

    Wang, T. C.; Liu, M. Y.; Zhu, J. Y.; Tao, A.; Kautz, J.; and Catanzaro, B. 2018 b . High-Resolution Image Synthesis and Semantic Manipulation with Conditional GANs. In IEEE/CVF Conference on Computer Vision and Pattern Recognition, 8798--8807. ISBN 2575-7075

    2018

  33. [41]

    Zhang, L.; Rao, A.; and Agrawala, M. 2023. Adding conditional control to text-to-image diffusion models. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 3836--3847

    2023

  34. [42]

    Zhang, P.; Yang, L.; Lai, J.; and Xie, X. 2022. Exploring Dual-task Correlation for Pose Guided Person Image Generation. In IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 7703--7712. ISBN 2575-7075

    2022

  35. [43]

    A.; Shechtman, E.; and Wang, O

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

    2018

  36. [44]

    C.; Sheikh, H

    Zhou, W.; Bovik, A. C.; Sheikh, H. R.; and Simoncelli, E. P. 2004. Image quality assessment: from error visibility to structural similarity. IEEE Transactions on Image Processing, 13(4): 600--612

    2004

  37. [45]

    Zhou, X.; Yin, M.; Chen, X.; Sun, L.; Gao, C.; and Li, Q. 2022. Cross Attention Based Style Distribution for Controllable Person Image Synthesis. In Computer Vision – ECCV 2022, 161--178. Springer Nature Switzerland. ISBN 978-3-031-19784-0

    2022

  38. [46]

    Zhou, X.; Zhang, B.; Zhang, T.; Zhang, P.; Bao, J.; Chen, D.; Zhang, Z.; and Wen, F. 2021. CoCosNet v2: Full-Resolution Correspondence Learning for Image Translation. In IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 11460--11470. ISBN 2575-7075

    2021

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.