Pith. sign in

REVIEW 4 major objections 6 minor 46 references

EOPose : Exemplar-based object reposing using Generalized Pose Correspondences

T0 review · 4 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 EOPose is a three-stage, class-agnostic pipeline that transfers the pose of a target product photo onto a source product photo using unsupervised keypoint correspondences, and it reports substantially better image-quality scores than the…

desk verdict A clean problem framing and a plausible three-stage pipeline, but the headline numbers in Table 1 cannot be pooled averages of the per-category rows, so the main quantitative claim needs checking before this is citable. read the letter →

arxiv 2505.03394 v1 pith:BQG65OLR submitted 2025-05-06 cs.CV

Sarthak Mehrotra , Rishabh Jain , Mayur Hemani , Balaji Krishnamurthy , Mausoom Sarkar This is my paper · ORCID
classification cs.CV
keywords objectreposingexemplar-basedimagesynthesisself-supervisedkeypointcorrespondencesDINO-ViTdensedescriptorsGANgenerationObjaversedatasetposetransfere-commerceproductimaging
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

EOPose claims that generic object reposing — changing a product photo into a new pose — can be done without per-class pose models by using dense visual correspondences from a pretrained self-supervised vision transformer. The paper proposes to find 35 matching keypoints between the appearance image and a pose-guidance image, warp the appearance image with a learned dense flow field, then re-render the warped image with a GAN that keeps fine-grained detail such as colors, textures, and brand marks. On a new Objaverse-based dataset of rendered shoes, vases, briefcases, and file cabinets, EOPose reports SSIM 0.44, LPIPS 0.33, and FID 18.22, ahead of the compared baselines, and a user study prefers it 88.4% of the time. If correct, the method offers e-commerce a way to produce multiple product views from two ordinary photos, without 3D models or manual editing.

What carries the argument

The load-bearing object is the generalized pose correspondence: 35 ordered keypoints per image found by extracting spatial descriptors from a pre-trained DINO-ViT, keeping only mutual nearest-neighbour (Best-Buddies) pairs between the appearance image and the pose image, and clustering the descriptor set with k-means. Each keypoint is encoded as a Gaussian heatmap channel, so the warping U-Net receives a 76-channel input (two RGB images plus 35 heatmaps per image) and predicts a coarse-to-fine flow pyramid; the flow is refined with convex upsampling and used to warp the source image. The final generator fuses pose and texture embeddings through texture injection blocks (scaling and shifting pose features by texture-derived parameters, followed by noise modulation) before tRGB blocks synthesize the output. This three-step design is what, in the paper's argument, preserves exact colors and brand marks while still transferring the new pose.

What would settle it

Randomly permute or zero the 70 Gaussian keypoint heatmap channels in the warping module's input while keeping the appearance and pose images unchanged; if the reported SSIM, LPIPS, and FID do not substantially worsen toward the TPS baseline, the keypoint correspondences are not the load-bearing component the paper claims.

Watch

Extended reading notes

Core claim

EOPose's central claim is that exemplar-based reposing of arbitrary object classes is achievable by replacing class-specific pose representations with unsupervised keypoint correspondences derived from DINO-ViT dense descriptors, matched by mutually-nearest-neighbour Best-Buddies Pairs and reduced to k=35 points by k-means clustering. These correspondences, rendered as Gaussian heatmap channels, guide a U-Net that predicts a pyramid of dense flow fields; the coarsest flow warps the appearance image, and intermediate flows are supervised with L1, perceptual, style, total-variation, and TPS-based losses. A StyleGAN-style generator then re-renders the warped image, injecting multi-scale texture encodings into pose embeddings through 2D style modulation and refining occluded or missing regions. The trained pipeline is evaluated on a new paired dataset built from 60 manually filtered Objaverse models rotated by random Euler angles, and the paper reports that EOPose achieves SSIM 0.44, LPIPS 0.33, and FID 18.22, substantially better than the next best values of SSIM 0.34, LPIPS 0.58, and FID 75.22 from the compared baselines.

Load-bearing premise

The pipeline's correctness rests on the assumption that the 35 correspondence points found by DINO-ViT and Best-Buddies Pairs stay semantically consistent across different instances and large pose changes, an assumption the paper never isolates by ablating the keypoint channels.

Editorial extensions

If this is right

  • For e-commerce settings where a brand has no 3D models, two ordinary photos of a same-class object suffice to synthesize any desired pose.
  • Because the first two stages warp rather than regenerate, the output preserves stickers, logos, patterns, and sharp color transitions that generative and diffusion models tend to hallucinate away.
  • The approach generalizes across object classes without retraining pose representations, provided the class has enough DINO-ViT-correspondence structure, as demonstrated on vases, briefcases, file cabinets, and shoes.
  • End-to-end fine-tuning after stage-wise training is essential: it improves FID from 34.86 to 18.22, so the two-stage training scheme alone is not enough for the reported quality.
  • The number of correspondence points has a sweet spot: 35 points outperform 15, 25, and 45, with FID degrading from 18.22 to 21.62, 20.65, and 22.48 respectively, showing that both too few and too many points hurt the flow quality.

Reading between the lines

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

  • A direct test the paper leaves open: ablate the keypoint channels; if EOPose's quality holds with random or zeroed keypoints, the claimed role of 'generalized pose correspondences' would be unsupported.
  • DINO-ViT correspondences were trained on natural images, and the four rendered Objaverse classes are relatively rigid, so applying EOPose to articulated or deformable objects would likely require denser or semantically finer correspondences than k=35.
  • Since the dataset is rendered from 3D models with controlled backgrounds, transferring the pipeline to real product photos will need extra robustness steps such as background removal, scale normalization, and camera calibration, which the paper's metrics do not measure.
  • The pipeline is a lightweight alternative to diffusion for controlled product-shot generation: it needs only a U-Net and a GAN, both far cheaper to run than iterative denoising, at the cost of relying on the source image to contain the appearance to preserve.
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

4 major / 6 minor

Summary. The paper introduces EOPose, an end-to-end exemplar-based object reposing method. Given an appearance image and a pose-guidance image of another object from the same class, EOPose first extracts sparse correspondences using DINO-ViT descriptors with Best-Buddies Pairs and k-means clustering, then warps the appearance image with a U-Net that predicts multi-scale flow fields, and finally re-renders the warped image with a GAN-based generator that injects multi-scale texture features into pose encodings. The authors also contribute a new paired dataset rendered from Objaverse models across four object categories and report quantitative results (SSIM, LPIPS, FID), qualitative comparisons, ablations, and a user study. The central claim is that EOPose substantially outperforms TPS warping, UFO-PT, and ControlCom on this dataset.

Significance. If the reported results are correct, the paper would make a useful practical contribution: it addresses a clear e-commerce need, proposes a category-agnostic pose representation via pretrained visual descriptors, and includes a new dataset that could enable further work on generic object reposing. The three-stage design is reasonable, and the ablations support the importance of end-to-end fine-tuning and the number of keypoints. The paper also ships several strengths: a concrete dataset construction protocol, a user study, and comparisons with multiple adapted baselines. However, the central quantitative evidence is currently undermined by an internally inconsistent aggregated row in Table 1 and by missing supplementary material that is explicitly referenced for key loss definitions and training details. These issues must be resolved before the comparative claims can be accepted.

major comments (4)
  1. [Section 5, Table 1] The 'All together' row cannot be a simple pooled mean of the four category rows for SSIM and LPIPS. For EOPose, the category SSIM values are 0.89, 0.86, 0.82, and 0.77, so any weighted average of per-image SSIM values over these disjoint groups must lie between 0.77 and 0.89; the reported 0.44 is outside this range. Likewise, the category LPIPS values 0.04, 0.06, 0.07, and 0.08 cannot average to 0.33. The same impossibility holds for the TPS row. Please state exactly how the 'All together' row was computed, whether it uses the same test samples as the category rows, and recompute all reported pooled values. As presented, the headline comparison 'EOPose achieves SSIM 0.44, LPIPS 0.33, FID 18.22 versus next-best 0.34, 0.58, 75.22' is not verifiable from the table.
  2. [Sections 3.2, 3.3, and 3.4] Several load-bearing definitions are deferred to a supplementary document that is not included in the arXiv submission: the TPS-based flow loss L_f low and the full form of Equation (3), the training schedule and epoch counts, all loss-weight hyperparameters (beta1...beta5, alpha_l1, alpha_per, alpha_sty, alpha_adv, alpha1, alpha2), and the dataset-split details. Without these, Equations (2) and (3) are not fully specified and the method cannot be reproduced. Please either include the supplementary material with the submission or move the essential definitions into the main text.
  3. [Section 3.1 and Table 2] The paper's central novelty is the use of unsupervised DINO-ViT keypoint correspondences for class-agnostic pose transfer, but no ablation removes or corrupts the keypoint channels. The only related experiment varies the number of keypoints k (15, 25, 35, 45); this does not test whether the correspondences are necessary, because the warping network also receives both original images and could in principle learn pose alignment from the images alone. Please add an ablation in which the Pa and Pp channels are removed or replaced with random keypoints, and report the resulting metrics. This is needed to support the claim that the correspondence module is an essential component.
  4. [Section 4, Baselines] The baseline comparison is not fully specified. The paper says UFO-PT and ControlCom are 'fine-tuned using our dataset' and that the TPS baseline is 'adapted from [39]', but it does not report the fine-tuning protocol (number of steps, learning rate, loss weighting, or model selection) or the TPS adaptation details. Without this information, the reader cannot assess whether the baselines were given a fair comparison; the reported margins may partly reflect suboptimal baseline tuning. Please provide the complete baseline training and evaluation protocol.
minor comments (6)
  1. [Abstract and Section 4] The abstract and Section 4 mention PSNR as an evaluation metric, but Table 1 and the experimental section report only SSIM, LPIPS, and FID. Please either add PSNR results or remove the mention.
  2. [Section 3.3, Equation (5)] Equation (5) appears to have the assignment direction reversed: the text states that the output e_i_out undergoes a noise modulation operation B to produce e_i_infused, but the equation reads e_i_out = B(e_i_infused). Please correct the notation so the text and equation agree.
  3. [Section 3.2, Equation (2)] The summation in Equation (2) is typeset incorrectly; the limits 'l=K-1 ... l=0' are ambiguous. Please write the summation with clear lower and upper limits, such as sum_{l=0}^{K-1}.
  4. [Section 6.2] The phrase 'try-on output' is used to describe the reposing result; for consistency with the rest of the paper, use 'reposing output'.
  5. [Table 2] The '35 keypoints' row and the 'EOPose (OURS)' row report identical numbers; please label the 35-keypoint row clearly as the full model rather than as an ablation, or explain the difference.
  6. [Section 5, User Study] The user study reports an 88.4% preference rate but provides no confidence interval or significance test; please report the number of comparisons per participant and the variance across participants.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: EOPose's empirical pipeline is self-contained and benchmarked against held-out ground-truth renderings.

full rationale

EOPose is an empirical systems paper rather than a derivation whose output reduces to its input. The claimed contribution is a three-stage reposing pipeline: DINO-ViT keypoint correspondences (Eq. 1), a learned warping U-Net (Eq. 2), and a GAN-based re-rendering generator (Eq. 6). Each stage consumes inputs that are distinct from the supervised targets: correspondences are computed from pre-trained DINO-ViT descriptors, warping is trained with L1/perceptual/style/flow losses against the ground-truth warped render, and the generator is trained with L1/perceptual/style/adversarial losses against the ground-truth final render. The reported SSIM/LPIPS/FID numbers compare generated images to held-out ground-truth renderings; no evaluation constant is fitted, and no headline metric is defined in terms of a method output that was also used to fit the model. Self-citations in the related-work section (ZFlow [7], VGFlow [17], UMFuse [16], SieveNet [18]) describe prior virtual try-on systems and are not the load-bearing evidence for EOPose's quantitative claims. The internal inconsistency in Table 1's 'All together' row relative to the category rows is a reporting/protocol concern, not a circular step, because it does not correspond to any equation or fitted parameter in the method. Under the rule that non-circular empirical benchmarking against external ground truth is scored in the 0-2 band, this paper receives 0.

Assumptions & free parameters 4 free parameters · 4 assumptions · 0 invented entities

The method is a learned pipeline; the central empirical result depends on hand-chosen hyperparameters (keypoint count, loss weights, pose range) and on assumptions about the transferability of DINO-ViT correspondences and synthetic training data to real objects. These are itemized above.

free parameters (4)
  • k (number of keypoints) = 35
    Set by hand for all experiments; the ablation in Table 2 shows SSIM varies from 0.42 to 0.45 and FID from 18.22 to 22.48 across k values, so the central result depends on this choice.
  • Loss weights (beta1..beta5, alpha_l1, alpha_per, alpha_sty, alpha_adv, alpha1, alpha2)
    Referred to as scalars in Equations 3, 6, and 7, but their values are not reported in the paper or in the absent supplementary; training balance depends on them.
  • Euler angle range for dataset rendering = 30 to 180 degrees
    Defines the pose distribution in the synthetic dataset; generalization outside this range is not tested.
  • Top-k selected patches = 35
    Patches are ranked by salience and the top-k are selected, tied to the k=35 keypoints used throughout.
assumptions (4)
  • domain assumption DINO-ViT features are semantically consistent across different instances of the same object class under large pose changes.
    The whole pipeline relies on these features for computing Pa and Pp (Section 3.1).
  • domain assumption Random Euler rotations of 3D models from Objaverse produce training pairs that approximate real object reposing conditions.
    The dataset construction (Section 3.4.1) assumes synthetic renderings transfer to the claimed e-commerce use case.
  • ad hoc to paper Losses L1, perceptual, style, and LSGAN are sufficient for faithful re-rendering that preserves textures and avoids hallucinations.
    The design of the generator and losses (Equation 6) is justified only by ablations on the synthetic dataset.
  • standard math The evaluation metrics SSIM, LPIPS, and FID are valid measures of reposing quality.
    Used in Section 4 without further justification; these are standard image generation metrics.

how reviews work

0 comments
Cite this review

Pith. "Pith review of EOPose : Exemplar-based object reposing using Generalized Pose Correspondences." pith.science (2026). https://pith.science/paper/BQG65OLR

@misc{pith2026250503394,
  author       = {Pith},
  title        = {Pith review of: EOPose : Exemplar-based object reposing using Generalized Pose Correspondences},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/BQG65OLR}},
  note         = {Machine review of arXiv:2505.03394}
}
read the original abstract

Reposing objects in images has a myriad of applications, especially for e-commerce where several variants of product images need to be produced quickly. In this work, we leverage the recent advances in unsupervised keypoint correspondence detection between different object images of the same class to propose an end-to-end framework for generic object reposing. Our method, EOPose, takes a target pose-guidance image as input and uses its keypoint correspondence with the source object image to warp and re-render the latter into the target pose using a novel three-step approach. Unlike generative approaches, our method also preserves the fine-grained details of the object such as its exact colors, textures, and brand marks. We also prepare a new dataset of paired objects based on the Objaverse dataset to train and test our network. EOPose produces high-quality reposing output as evidenced by different image quality metrics (PSNR, SSIM and FID). Besides a description of the method and the dataset, the paper also includes detailed ablation and user studies to indicate the efficacy of the proposed method

Figures

Figures reproduced from arXiv: 2505.03394 by the authors.

Figure 1
Figure 1. Exemplar-based object reposing involves synthesizing [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. A) Keypoint Detection: This module uses a pre-trained DINO-ViT to detect visual correspondences between the given pose [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Our Generator module consumes the warped image [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: Paired dataset preparation by rendering Objaverse[ [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]
Figure 5
Figure 5. Figure 5: In this figure we show improvements along differ [PITH_FULL_IMAGE:figures/full_fig_p007_5.png]
Figure 6
Figure 6. Figure 6: Survey results indicating the percentage of images where [PITH_FULL_IMAGE:figures/full_fig_p007_6.png]
Figure 7
Figure 7. Figure 7: Comparison of images after warping using flows from [PITH_FULL_IMAGE:figures/full_fig_p008_7.png]
Figure 8
Figure 8. Figure 8: This figure illustrates the importance of the pose image [PITH_FULL_IMAGE:figures/full_fig_p008_8.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

46 extracted references · 35 canonical work pages

  1. [39]

    Toward characteristic- preserving image-based virtual try-on network

    Bochao Wang, Huabin Zheng, Xiaodan Liang, Yimin Chen, Liang Lin, and Meng Yang. Toward characteristic- preserving image-based virtual try-on network. In Pro- ceedings of the European Conference on Computer Vision (ECCV), pages 589–604, 2018. 1, 2, 6

    work page 2018

  2. [1]

    Pose with Style: Detail- preserving pose-guided image synthesis with conditional stylegan

    Badour AlBahar, Jingwan Lu, Jimei Yang, Zhixin Shu, Eli Shechtman, and Jia-Bin Huang. Pose with Style: Detail- preserving pose-guided image synthesis with conditional stylegan. ACM Transactions on Graphics, 2021. 5

    work page 2021

  3. [2]

    Deep vit features as dense visual descriptors

    Shir Amir, Yossi Gandelsman, Shai Bagon, and Tali Dekel. Deep vit features as dense visual descriptors. ECCVW What is Motion For?, 2022. 3

    work page 2022

  4. [3]

    A note on the inception score

    Shane Barratt and Rishi Sharma. A note on the inception score. arXiv preprint arXiv:1801.01973, 2018. 6

    arXiv 2018

  5. [4]

    Lightplane: Highly-scalable components for neu- ral 3d fields

    Ang Cao, Justin Johnson, Andrea Vedaldi, and David Novotny. Lightplane: Highly-scalable components for neu- ral 3d fields. arXiv, 2024. 1

    work page 2024

  6. [5]

    Realtime multi-person 2d pose estimation using part affinity fields

    Zhe Cao, Tomas Simon, Shih-En Wei, and Yaser Sheikh. Realtime multi-person 2d pose estimation using part affinity fields. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017. 1

    work page 2017

  7. [6]

    Learning continuous 3d words for text-to- image generation

    Ta-Ying Cheng, Matheus Gadelha, Thibault Groueix, Matthew Fisher, Radomir Mech, Andrew Markham, and Niki Trigoni. Learning continuous 3d words for text-to- image generation. 2024. 1

    work page 2024

  8. [7]

    Zflow: Gated appearance flow-based virtual try-on with 3d priors

    Ayush Chopra, Rishabh Jain, Mayur Hemani, and Balaji Kr- ishnamurthy. Zflow: Gated appearance flow-based virtual try-on with 3d priors. In Proceedings of the IEEE/CVF In- ternational Conference on Computer Vision (ICCV) , pages 5433–5442, 2021. 1, 2

    work page 2021

Show all 46 references
  1. [8]

    Obja- verse: A universe of annotated 3d objects

    Matt Deitke, Dustin Schwenk, Jordi Salvador, Luca Weihs, Oscar Michel, Eli VanderBilt, Ludwig Schmidt, Kiana Ehsani, Aniruddha Kembhavi, and Ali Farhadi. Obja- verse: A universe of annotated 3d objects. arXiv preprint arXiv:2212.08051, 2022. 5, 6

    2022 arXiv

  2. [9]

    Alpha- pose: Whole-body regional multi-person pose estimation and tracking in real-time

    Hao-Shu Fang, Jiefeng Li, Hongyang Tang, Chao Xu, Haoyi Zhu, Yuliang Xiu, Yong-Lu Li, and Cewu Lu. Alpha- pose: Whole-body regional multi-person pose estimation and tracking in real-time. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022. 1

    2022

  3. [10]

    End-to-end learning of keypoint detector and descriptor for pose invariant 3d matching

    Georgios Georgakis, Srikrishna Karanam, Ziyan Wu, Jan Ernst, and Jana Ko ˇseck´a. End-to-end learning of keypoint detector and descriptor for pose invariant 3d matching. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2018. 2

    2018

  4. [11]

    Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courville, and Yoshua Bengio

    Ian J. Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courville, and Yoshua Bengio. Generative adversarial networks, 2014. 4

    2014

  5. [12]

    Densepose: Dense human pose estimation in the wild

    Rıza Alp G ¨uler, Natalia Neverova, and Iasonas Kokkinos. Densepose: Dense human pose estimation in the wild. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pages 7297–7306, 2018. 1, 2

    2018

  6. [13]

    Viton: An image-based virtual try-on network

    Xintong Han, Zuxuan Wu, Zhe Wu, Ruichi Yu, and Larry S Davis. Viton: An image-based virtual try-on network. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 7543–7552, 2018. 1

    2018

  7. [14]

    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. In Advances in neural information processing systems, pages 6626–6637, 2017. 6

    2017

  8. [15]

    3d human body reconstruction from a single image via volumetric regression

    Aaron S Jackson, Chris Manafas, and Georgios Tzimiropou- los. 3d human body reconstruction from a single image via volumetric regression. In Proceedings of the European Con- ference on Computer Vision (ECCV), pages 0–0, 2018. 2

    2018

  9. [16]

    Umfuse: Unified multi view fusion for human editing applications, 2023

    Rishabh Jain, Mayur Hemani, Duygu Ceylan, Krishna Ku- mar Singh, Jingwan Lu, Mausoom Sarkar, and Balaji Krish- namurthy. Umfuse: Unified multi view fusion for human editing applications, 2023. 1, 2

    2023

  10. [17]

    Vgflow: Visibility guided flow network for hu- man reposing

    Rishabh Jain, Krishna Kumar Singh, Mayur Hemani, Jing- wan Lu, Mausoom Sarkar, Duygu Ceylan, and Balaji Krish- namurthy. Vgflow: Visibility guided flow network for hu- man reposing. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pa...

    2023

  11. [18]

    Sievenet: A unified framework for robust image-based virtual try-on

    Surgan Jandial, Ayush Chopra, Kumar Ayush, Mayur He- mani, Balaji Krishnamurthy, and Abhijeet Halwai. Sievenet: A unified framework for robust image-based virtual try-on. In The IEEE Winter Conference on Applications of Computer Vision, pages 2182–2190, 2020. 2

    2020

  12. [19]

    Stableviton: Learning semantic correspon- dence with latent diffusion model for virtual try-on, 2023

    Jeongho Kim, Gyojung Gu, Minho Park, Sunghyun Park, and Jaegul Choo. Stableviton: Learning semantic correspon- dence with latent diffusion model for virtual try-on, 2023. 2

    2023

  13. [20]

    Adam: A method for stochastic optimization

    Diederik P Kingma and Jimmy Ba. Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980 ,

    arXiv

  14. [21]

    Pona: Pose-guided non-local attention for human pose transfer

    Kun Li, Jinsong Zhang, Yebin Liu, Yu-Kun Lai, and Qiong- hai Dai. Pona: Pose-guided non-local attention for human pose transfer. IEEE Transactions on Image Processing, 29: 9584–9599, 2020. 1

    2020

  15. [22]

    Dense in- trinsic appearance flow for human pose transfer

    Yining Li, Chen Huang, and Chen Change Loy. Dense in- trinsic appearance flow for human pose transfer. InProceed- ings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 3693–3702, 2019

    2019

  16. [23]

    Neural actor: Neural free-view synthesis of human actors with pose con- trol

    Lingjie Liu, Marc Habermann, Viktor Rudnev, Kripasindhu Sarkar, Jiatao Gu, and Christian Theobalt. Neural actor: Neural free-view synthesis of human actors with pose con- trol. ACM transactions on graphics (TOG) , 40(6):1–16,

  17. [24]

    Smpl: A skinned multi- person linear model

    Matthew Loper, Naureen Mahmood, Javier Romero, Gerard Pons-Moll, and Michael J Black. Smpl: A skinned multi- person linear model. ACM transactions on graphics (TOG), 34(6):1–16, 2015. 1, 2

    2015

  18. [25]

    Object recognition from local scale-invariant features

    David G Lowe. Object recognition from local scale-invariant features. In Proceedings of the seventh IEEE interna- tional conference on computer vision , pages 1150–1157. Ieee, 1999. 2

    1999

  19. [26]

    Lau, Zhen Wang, and Stephen Paul Smolley

    Xudong Mao, Qing Li, Haoran Xie, Raymond Y .K. Lau, Zhen Wang, and Stephen Paul Smolley. Least squares gen- erative adversarial networks. In Proceedings of the IEEE International Conference on Computer Vision (ICCV), 2017. 5

    2017

  20. [27]

    Freeman, and Shai Avidan

    Shaul Oron, Tali Dekel, Tianfan Xue, William T. Freeman, and Shai Avidan. Best-buddies similarity - robust template matching using mutual nearest neighbors, 2016. 3

    2016

  21. [28]

    Scalable diffusion models with transformers, 2023

    William Peebles and Saining Xie. Scalable diffusion models with transformers, 2023. 2 9

    2023

  22. [29]

    Clothcap: Seamless 4d clothing capture and retar- geting

    Gerard Pons-Moll, Sergi Pujades, Sonny Hu, and Michael J Black. Clothcap: Seamless 4d clothing capture and retar- geting. ACM Transactions on Graphics (TOG), 36(4):1–15,

  23. [30]

    Learning transferable visual models from natural language supervision, 2021

    Alec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh, Gabriel Goh, Sandhini Agarwal, Girish Sastry, Amanda Askell, Pamela Mishkin, Jack Clark, Gretchen Krueger, and Ilya Sutskever. Learning transferable visual models from natural language supervision, 2021. 2

    2021

  24. [31]

    U-net: Convolutional networks for biomedical image segmentation

    Olaf Ronneberger, Philipp Fischer, and Thomas Brox. U-net: Convolutional networks for biomedical image segmentation. CoRR, abs/1505.04597, 2015. 4

    2015 arXiv

  25. [32]

    Virtual fitting by single-shot body shape estimation

    Masahiro Sekine, Kaoru Sugita, Frank Perbet, Bj¨orn Stenger, and Masashi Nishiyama. Virtual fitting by single-shot body shape estimation. In Int. Conf. on 3D Body Scanning Tech- nologies, pages 406–413. Citeseer, 2014. 2

    2014

  26. [33]

    Unifying analy- sis of full reference image quality assessment

    Kalpana Seshadrinathan and Alan C Bovik. Unifying analy- sis of full reference image quality assessment. In 2008 15th IEEE International Conference on Image Processing, pages 1200–1203. IEEE, 2008. 6

    2008

  27. [34]

    Object- stitch: Object compositing with diffusion model

    Yizhi Song, Zhifei Zhang, Zhe Lin, Scott Cohen, Brian Price, Jianming Zhang, Soo Ye Kim, and Daniel Aliaga. Object- stitch: Object compositing with diffusion model. In Pro- ceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 18310–18319...

    2023

  28. [35]

    Imprint: Generative object compositing by learning identity-preserving representation

    Yizhi Song, Zhifei Zhang, Zhe Lin, Scott Cohen, Brian Price, Jianming Zhang, Soo Ye Kim, He Zhang, Wei Xiong, and Daniel Aliaga. Imprint: Generative object compositing by learning identity-preserving representation. In Proceedings of the IEEE/CVF Conference on Computer Vision ...

    2024

  29. [36]

    General object pose transformation network from unpaired data

    Yukun Su, Guosheng Lin, Ruizhou Sun, and Qingyao Wu. General object pose transformation network from unpaired data. In European Conference on Computer Vision , pages 292–310. Springer, 2022. 2, 6, 7

    2022

  30. [37]

    Raft: Recurrent all-pairs field transforms for optical flow, 2020

    Zachary Teed and Jia Deng. Raft: Recurrent all-pairs field transforms for optical flow, 2020. 4

    2020

  31. [38]

    Bodynet: V olu- metric inference of 3d human body shapes

    Gul Varol, Duygu Ceylan, Bryan Russell, Jimei Yang, Ersin Yumer, Ivan Laptev, and Cordelia Schmid. Bodynet: V olu- metric inference of 3d human body shapes. In Proceedings of the European Conference on Computer Vision (ECCV) ,

  32. [40]

    Mv-vton: Multi-view virtual try-on with diffusion models

    Haoyu Wang, Zhilu Zhang, Donglin Di, Shiliang Zhang, and Wangmeng Zuo. Mv-vton: Multi-view virtual try-on with diffusion models. arXiv preprint arXiv:2404.17364, 2024. 2

    2024 arXiv

  33. [41]

    Effec- tive whole-body pose estimation with two-stages distillation

    Zhendong Yang, Ailing Zeng, Chun Yuan, and Yu Li. Effec- tive whole-body pose estimation with two-stages distillation. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 4210–4220, 2023. 1

    2023

  34. [42]

    Controlcom: Controllable im- age composition using diffusion model, 2023

    Bo Zhang, Yuxuan Duan, Jun Lan, Yan Hong, Huijia Zhu, Weiqiang Wang, and Li Niu. Controlcom: Controllable im- age composition using diffusion model, 2023. 1, 2, 6, 7, 8

    2023

  35. [43]

    Efros, Eli Shecht- man, and Oliver Wang

    Richard Zhang, Phillip Isola, Alexei A. Efros, Eli Shecht- man, and Oliver Wang. The unreasonable effectiveness of deep features as a perceptual metric, 2018. 6

    2018

  36. [44]

    Deephuman: 3d human reconstruction from a single image

    Zerong Zheng, Tao Yu, Yixuan Wei, Qionghai Dai, and Yebin Liu. Deephuman: 3d human reconstruction from a single image. In Proceedings of the IEEE International Con- ference on Computer Vision, pages 7739–7749, 2019. 2

    2019

  37. [45]

    Starmap for category-agnostic keypoint and viewpoint esti- mation

    Xingyi Zhou, Arjun Karpur, Linjie Luo, and Qixing Huang. Starmap for category-agnostic keypoint and viewpoint esti- mation. In Proceedings of the European Conference on Com- puter Vision (ECCV), 2018. 1, 2

    2018

  38. [46]

    M&m vto: Multi- garment virtual try-on and editing

    Luyang Zhu, Yingwei Li, Nan Liu, Hao Peng, Dawei Yang, and Ira Kemelmacher-Shlizerman. M&m vto: Multi- garment virtual try-on and editing. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2024. 2 10

    2024

Pith tools

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