Pith. sign in

REVIEW 3 major objections 5 minor 39 references

Adapting Human Mesh Recovery with Vision-Language Feedback

T0 review · 3 major / 5 minor · reviewed 2026-08-09 · deepseek-v4-flash

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

Pith's one-line read The paper claims that fine-grained textual pose descriptions, aligned with a pose VQ-VAE and used as diffusion guidance, improve monocular human mesh recovery.

desk verdict A reasonable text-guided diffusion refinement for HMR whose central claim—that VLM text semantics improve 3D pose—is plausible but not yet separated from regularization noise. read the letter →

arxiv 2502.03836 v1 pith:RHACHDTH submitted 2025-02-06 cs.CV

Chongyang Xu , Buzhen Huang , Chengfang Zhang , Ziliang Feng , Yangang Wang This is my paper · ORCID
classification cs.CV
keywords humanmeshrecoveryvision-languagefeedbackdiffusionguidanceVQ-VAEtext-posealignmentSMPLmulti-modalsignal3Dposeestimation
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

Human mesh recovery from a single image usually means choosing between regression models, which get joint positions right but fit the image poorly, and optimization methods, which align the image well but suffer from depth ambiguity. This paper tries to get both by adding a third signal: fine-grained textual descriptions of body parts and their interactions, generated by a vision-language model. The central claim is that those text descriptions carry 3D pose information that can constrain the optimization and reduce depth ambiguity. The method aligns text and pose in a shared discrete latent space, then uses a diffusion process to refine an initial regression estimate under gradients from 2D keypoints and from the text-pose cosine similarity. On the standard 3DPW and Human3.6M benchmarks, the refinement reaches 43.9 mm and 32.4 mm in PA-MPJPE (pose error after rigid alignment), better than HMR2.0 by 0.4 mm and 1.2 mm.

What carries the argument

The central machinery is a shared discrete latent space between pose and text. A pose VQ-VAE quantizes body poses into codebook tokens, and a text encoder maps CLIP text embeddings into the same space via contrastive learning plus a reconstruction loss $L_{\mathrm{rec}} = \lVert D_p(E_t(f_c)) - \theta\rVert^2$. The result is a differentiable text-pose similarity $L_{\mathrm{cos}}$ (Eq. 7), whose gradient with respect to the pose parameters, $G_{\mathrm{text}} = \partial L_{\mathrm{cos}}/\partial \hat{\theta}$, joins the 2D keypoint reprojection gradient and the image features to form the condition for the diffusion denoiser. The diffusion process treats the initial regression as the mean of a Gaussian distribution and uses the conditional gradient to push the sample toward the true pose distribution.

What would settle it

A controlled ablation on 3DPW that replaces ChatPose descriptions with unrelated or scrambled text while keeping image features and keypoint gradients identical would settle whether the text modality contributes independent 3D signal: if PA-MPJPE stays the same as the full method, the text branch is not doing the claimed work.

Watch

Extended reading notes

Core claim

On the paper's own account, the discovery it establishes is that fine-grained interactive textual descriptions can enhance human mesh recovery. The method couples a ViT-based regressor that predicts initial SMPL parameters with a pose VQ-VAE and a CLIP-derived text encoder, trained together so that text features and pose codes live in a shared latent space. During inference, a diffusion model refines the initial pose under three conditions: image features from the backbone, gradients of the 2D keypoint reprojection error, and gradients of the cosine similarity $L_{\mathrm{cos}}$ between the pose code and the text description. With all three conditions, the method reports 43.9 mm PA-MPJPE on 3DPW and 32.4 mm on Human3.6M, improving on HMR2.0 by 0.4 mm and 1.2 mm respectively.

Load-bearing premise

The text-guidance effect rests on the assumption that the vision-language model's descriptions encode 3D pose information beyond what the image already shows, so the gradient of the learned cosine similarity points toward the true 3D pose in the discrete latent space for unseen images.

Editorial extensions

If this is right

  • With all three conditions active, the method improves PA-MPJPE on 3DPW and Human3.6M over its regression baseline and over HMR2.0, indicating that multi-modal feedback refines 3D pose beyond what a single modality provides.
  • The text modality is most useful when 2D keypoints are noisy or ambiguous, since the ablations show the largest contribution from keypoints and the best final result when text is added.
  • Because the same latent alignment pipeline accepts any vision-language model that can describe body poses, the framework can absorb future, more capable VLMs without changing the optimization structure.
  • The reported results establish a new reference point for regression-plus-optimization human mesh recovery, and the code release allows direct comparison with the method.

Reading between the lines

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

  • An experiment the paper leaves implicit: substitute randomized or unrelated text descriptions while keeping image and keypoint guidance identical. If PA-MPJPE stays unchanged, the text branch is not contributing independent 3D signal; if it degrades, the claim is supported.
  • If the text-pose alignment genuinely encodes 3D structure, the same shared latent space could support zero-shot text-driven pose editing or natural-language pose retrieval, directions the paper does not discuss.
  • A hidden risk is the discrete VQ-VAE codebook: rare or extreme poses may quantize to only a few tokens, so the cosine gradient could pull toward an average pose rather than the true one. This could be tested by measuring error on hard poses or unseen viewpoint variations.
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

3 major / 5 minor

Summary. The paper proposes a diffusion-based refinement framework for monocular human mesh recovery. Starting from an HMR2.0-style ViT regressor, it runs a conditional diffusion process whose guidance combines image features, gradients from 2D keypoint reprojection, and gradients from a text-pose cosine similarity. The text is obtained by prompting ChatPose to describe body parts, and a VQ-VAE plus contrastive text encoder is trained to align text and pose latents. Experiments on 3DPW and Human3.6M show PA-MPJPE improvements over HMR2.0 of 0.6 and 1.2 mm, respectively, and ablations attribute roughly 0.4 to 0.6 mm PA-MPJPE to the text condition. The central claim is that fine-grained visual-language descriptions enhance human mesh recovery.

Significance. If the central claim were established, the use of VLM-generated text as an optimization-side constraint would be a useful addition to the human mesh recovery literature, connecting recent work on text-to-motion with regression-plus-diffusion refinement. The paper is clearly written and the idea is timely, and the authors are to be credited for explicitly ablating each condition and for committing to a public code release. However, the experimental support for the text-specific contribution is thin: the effect is under 1 mm PA-MPJPE with no variance estimates, and the raw MPJPE on Human3.6M is worse than the reported HMR2.0 backbone. The core update rule (Eq. 5) is also asserted without derivation. These are load-bearing gaps that prevent the paper from making its contribution credible as submitted.

major comments (3)
  1. [Section II-D1, Eq. (5)] The update rule Δx_t = Δt · ∇_x log q(x_t | c) is asserted without derivation or citation. The preceding score-matching objective (Eq. 4) is stated for an unconditional score s_model(x; ϕ) and does not imply this conditional form. No definition is given for q(x_t | c), no role for the timestep t is specified, and Δt is not characterized (learning rate, noise scale, or discretization step). Because this equation is the core of the proposed adaptation mechanism, the method is not reproducible as written. Please derive the rule from the score-matching loss or cite the exact guidance scheme, and specify how the conditions I, G_keyp, and G_text enter q(x_t | c).
  2. [Section III-C2, Table II] The central claim that textual descriptions enhance mesh recovery is not established. The gain from the text condition is 0.6 mm PA-MPJPE on 3DPW and 0.4 mm on Human3.6M when comparing 'w/ all conditions' with 'w/o text', and no standard deviations, repeated-seed runs, or significance tests are reported; these differences are within typical run-to-run noise for HMR systems. In addition, the 'w/ text' row (72.8/47.0 on 3DPW) is worse than 'w/ image' (70.6/45.6), which is consistent with the text channel acting as a weak auxiliary regularizer rather than as a source of independent 3D pose information. Because the descriptions are generated from the same input image, the text is a deterministic function of the image, and the contrastive alignment (Eq. 7) may simply map image-derived text to the latent neighborhood of the regressor's own predictions. To support contribution (2), the authors should add a shuffled-text or unrelated-text control while keeping image and keypoint conditions fixed, compare against a zero-gradient text channel, and report error bars over at least three independent runs.
  3. [Table I and Table II] On Human3.6M, the method's raw MPJPE (47.7 mm) is worse than the HMR2.0 backbone (44.8 mm) while only PA-MPJPE improves (32.4 vs 33.6 mm). Since the paper claims reduced depth ambiguity and accurate 3D pose perception, this degradation in unaligned MPJPE needs an explicit explanation. Moreover, the 'Initial Prediction' row in Table II (56.4/34.0 on Human3.6M and 73.4/47.5 on 3DPW) is substantially worse than the HMR2.0 numbers reported in Table I, despite using the same ViT-H backbone; the source of this discrepancy should be clarified because it changes the interpretation of the diffusion gains. If the benefit is largely in rigidly aligned error, the authors should state that and support the claim of 3D accuracy with appropriate metrics.
minor comments (5)
  1. [Section III-B] The sentence claiming the PA-MPJPE 'improves by 0.4' on 3DPW is inconsistent with Table I, where the improvement over HMR2.0 is 44.5 − 43.9 = 0.6 mm; please correct the number.
  2. [Eqs. (6) and (7)] In Eq. (7), L_cos is written as a squared cosine similarity between pose and text features, but G_text is defined as ∂L_cos/∂θ̂ while G_keyp in Eq. (6) is defined with respect to J3D; the paper should state the exact variable of differentiation and the dimensionality of each gradient, since c is a concatenation of both.
  3. [Section III-A2] Section III-A2 lists datasets, epochs, and GPUs but omits the diffusion hyperparameters (number of denoising steps K, initial standard deviation σ, gradient scaling factor Δt, and contrastive temperature τ); these are necessary for reproducibility and should be reported.
  4. [Supplementary Fig. 5] Figure 5 in the supplementary material contains garbled mixed-language text in the prompt examples; a clean English version should be provided.
  5. [Section II-D] Section II-D would benefit from an explicit comparison between the proposed 'distribution adaptation' and the score-guided diffusion approach of [13]; currently the relationship and the claimed novelty are not clearly delineated.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the text-guided diffusion is a learned auxiliary condition evaluated on external benchmarks, not a fitted quantity renamed as a prediction.

full rationale

The paper's derivation chain is not circular in the sense defined here. The text descriptions are indeed generated by ChatPose from the same input image (Sec. II-C1), and the text encoder is trained with the alignment loss L_align (Eq. 3), including a reconstruction term L_rec = ||D_p(E_t(f_c)) - θ||^2, so the text branch is a learned, image-derived pose prior rather than an independent measurement. However, the paper never equates the final output to that prior: the final pose is produced by a separately trained diffusion denoiser conditioned on concatenated image features, keypoint gradients, and text gradients (Sec. II-D), and the ablations show the text condition alone is far worse than the full condition (72.8/47.0 vs 69.3/43.9 on 3DPW). The benchmark numbers are external (3DPW, and Human3.6M under the standard protocol), so the result is not forced by the authors' own equations. The self-citations ([14], [15]) are used only to justify the initial-regressor-plus-diffusion recipe and are not load-bearing support for the text-feedback claim. Concerns that the text branch may add no semantic information beyond the image (e.g., lack of a shuffled-text control, no variance estimates) are experimental-evidence weaknesses, not circular reasoning: they do not reduce any prediction to a fit by construction. Accordingly, no circular step meets the evidentiary bar of quoting an equation-level equivalence or a fitted parameter renamed as a prediction.

Assumptions & free parameters 6 free parameters · 4 assumptions · 0 invented entities

The method has no new physical entities. Its scientific contribution rests on several learned components and hand-set hyperparameters; the diffusion guidance equation is asserted rather than derived, and the text signal is itself generated from the image, so it is not an independent measurement.

free parameters (6)
  • lambda_smpl, lambda_joint, lambda_reproj = not reported
    Loss weights in Eq. (1) controlling the regressor training; values are not given.
  • alpha (VQ-VAE commitment weight) = not reported
    Hyperparameter in Eq. (2) for pose codebook training; value is not given.
  • tau (contrastive temperature) = not reported
    Temperature in Eq. (3) scaling the contrastive logits; value is not given.
  • Delta_t (gradient scaling factor) = not reported
    Update step size in Eq. (5); no value or schedule is provided.
  • sigma (initial distribution std) = not reported
    Standard deviation of the assumed Gaussian around the initial prediction; value is not given.
  • diffusion step count K = not reported
    Number of denoising steps is shown as K in Fig. 3 but is not specified in the text.
assumptions (4)
  • domain assumption SMPL with 6D rotation representation adequately represents human pose and shape.
    Used throughout as the body parameterization in Section II-A.
  • domain assumption CLIP text embeddings provide a suitable base for aligning text to pose via contrastive learning.
    The text encoder starts from CLIP features in Section II-C2 and assumes they carry enough semantics for pose.
  • ad hoc to paper ChatPose descriptions of the image contain 3D pose information beyond what is visible in 2D.
    The whole method depends on this, stated in Section II-C and II-D2, but no experiment verifies the independence of text from image cues.
  • ad hoc to paper The diffusion update rule in Eq. (5), Delta_x_t = Delta_t * gradient of log q(x_t|c), is a valid posterior sampling step.
    This is the central guidance mechanism and is asserted without derivation in Section II-D1.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Adapting Human Mesh Recovery with Vision-Language Feedback." pith.science (2026). https://pith.science/paper/RHACHDTH

@misc{pith2026250203836,
  author       = {Pith},
  title        = {Pith review of: Adapting Human Mesh Recovery with Vision-Language Feedback},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/RHACHDTH}},
  note         = {Machine review of arXiv:2502.03836}
}
read the original abstract

Human mesh recovery can be approached using either regression-based or optimization-based methods. Regression models achieve high pose accuracy but struggle with model-to-image alignment due to the lack of explicit 2D-3D correspondences. In contrast, optimization-based methods align 3D models to 2D observations but are prone to local minima and depth ambiguity. In this work, we leverage large vision-language models (VLMs) to generate interactive body part descriptions, which serve as implicit constraints to enhance 3D perception and limit the optimization space. Specifically, we formulate monocular human mesh recovery as a distribution adaptation task by integrating both 2D observations and language descriptions. To bridge the gap between text and 3D pose signals, we first train a text encoder and a pose VQ-VAE, aligning texts to body poses in a shared latent space using contrastive learning. Subsequently, we employ a diffusion-based framework to refine the initial parameters guided by gradients derived from both 2D observations and text descriptions. Finally, the model can produce poses with accurate 3D perception and image consistency. Experimental results on multiple benchmarks validate its effectiveness. The code will be made publicly available.

Figures

Figures reproduced from arXiv: 2502.03836 by the authors.

Figure 1
Figure 1. (a) Regression-based methods struggle with model-image alignment for challenging poses. (b) Optimization-based methods are prone to overfitting noisy 2D inputs and suffer from severe depth ambiguity. (c) Our method leverages prior knowledge from large vision-language models to improve both 2D and 3D performance. parameters and then refine the results using additional obser￾vations, such as 2D keypoints [12] and phys… view at source ↗
Figure 2
Figure 2. Pose-Text Alignment. We first train a discrete pose codebook via VQ-VAE. To bridge the gap between text and 3D pose modalities, we then train a text encoder to align the texts to body poses in latent space with contrastive learning. C. Description Extraction and Modal Alignment Texts contain rich 3D information for describing human body poses, such as joint positions, part orientations and intra￾body interactions, w… view at source ↗
Figure 3
Figure 3. Overview of our method. Given an image, a large vision-language model is first used to extract detailed interactive descriptions for the body parts. An initial prediction is then made, followed by the construction of a diffusion-based framework that refines this prediction using multi-modal feedback. At each time step, the gradients of 2D keypoints are computed, along with the similarity loss between text embeddings… view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: Qualitative results. From left to right: RGB image, ProHMR [33], HMR2.0 [8], and our method. Our approach ensures accurate 3D joint positions with minimal depth ambiguity while achieving robust front-facing alignment. The gradient of the pose parameters with respect to…
Figure 6
Figure 6. Figure 6: Schematic of description generation and reorganization. We use the generated prompts to create descriptive texts for each image. The key information from these texts is then extracted, and the final description is truncated to 77 words or fewer. Self-attention Block Ad…
Figure 7
Figure 7. Figure 7: Architecture of our conditional diffusion model. We adopt the Transformer architecture and replace the standard normalization layer with an adaptive normalization layer. This layer combines the noisy SMPL parameters, positional embeddings, and observations through adap…
Figure 8
Figure 8. Figure 8: More qualitative results. From left to right are our method, HMR2, and ProHMR, including both front and side views. Our method has good alignment with better 3D accuracy [PITH_FULL_IMAGE:figures/full_fig_p006_8.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

39 extracted references · 39 canonical work pages

  1. [1]

    Hierarchical spatial-temporal adaptive graph fusion for monocular 3d human pose estimation,

    Lijun Zhang, Feng Lu, Kangkang Zhou, Xiang-Dong Zhou, and Yu Shi, “Hierarchical spatial-temporal adaptive graph fusion for monocular 3d human pose estimation,” IEEE Signal Processing Letters , vol. 31, pp. 61–65, 2024

    work page 2024

  2. [2]

    Segmentation-based background-inference and small-person pose es- timation,

    Fei Gao, Hua Li, Jiyou Fei, Yangjie Huang, and Long Liu, “Segmentation-based background-inference and small-person pose es- timation,” IEEE Signal Processing Letters , vol. 29, pp. 1584–1588, 2022

    work page 2022

  3. [3]

    Keypoint aware robust representation for transformer-based re- identification of occluded person,

    Sejun Kim, Sungjae Kang, Hyomin Choi, Seong Soo Kim, and Kisung Seo, “Keypoint aware robust representation for transformer-based re- identification of occluded person,” IEEE Signal Processing Letters , vol. 30, pp. 65–69, 2023

    work page 2023

  4. [4]

    Curricu- lum enhanced supervised attention network for person re-identification,

    Xiaoguang Zhu, Jiuchao Qian, Haoyu Wang, and Peilin Liu, “Curricu- lum enhanced supervised attention network for person re-identification,” IEEE Signal Processing Letters , vol. 27, pp. 1665–1669, 2020

    work page 2020

  5. [5]

    Pose attention-guided paired-images generation for visible-infrared person re-identification,

    Yongheng Qian and Su-Kit Tang, “Pose attention-guided paired-images generation for visible-infrared person re-identification,” IEEE Signal Processing Letters, vol. 31, pp. 346–350, 2024

    work page 2024

  6. [6]

    Exploiting attribute dependency for attribute assignment in crowded scenes,

    Chunhua Deng, Zhiguo Cao, Yang Xiao, Hao Lu, Ke Xian, and Yin Chen, “Exploiting attribute dependency for attribute assignment in crowded scenes,” IEEE Signal Processing Letters , vol. 23, no. 10, pp. 1325–1329, 2016

    work page 2016

  7. [7]

    End-to-end recovery of human shape and pose,

    Angjoo Kanazawa, Michael J. Black, David W. Jacobs, and Jitendra Malik, “End-to-end recovery of human shape and pose,” in Computer Vision and Pattern Recognition (CVPR) , 2018

    work page 2018

  8. [8]

    Humans in 4D: Reconstructing and tracking humans with transformers,

    Shubham Goel, Georgios Pavlakos, Jathushan Rajasegaran, Angjoo Kanazawa, and Jitendra Malik, “Humans in 4D: Reconstructing and tracking humans with transformers,” in ICCV, 2023

    work page 2023

Show all 39 references
  1. [9]

    Keep it SMPL: Automatic estimation of 3D human pose and shape from a single image,

    Federica Bogo, Angjoo Kanazawa, Christoph Lassner, Peter Gehler, Javier Romero, and Michael J. Black, “Keep it SMPL: Automatic estimation of 3D human pose and shape from a single image,” in Computer Vision – ECCV 2016 . Oct. 2016, Lecture Notes in Computer Science, Springer In...

    2016

  2. [10]

    Refit: Recurrent fitting network for 3d human recovery,

    Yufu Wang and Kostas Daniilidis, “Refit: Recurrent fitting network for 3d human recovery,” in International Conference on Computer Vision (ICCV), 2023

    2023

  3. [11]

    Alphapose: 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, “Alphapose: Whole-body regional multi-person pose estimation and tracking in real-time,” IEEE Transactions on Pattern Analysis and Machine Intelligence , vol. 45, no. 6, pp. 7157...

    2022

  4. [12]

    Learning to reconstruct 3d human pose and shape via model-fitting in the loop,

    Nikos Kolotouros, Georgios Pavlakos, Michael J Black, and Kostas Daniilidis, “Learning to reconstruct 3d human pose and shape via model-fitting in the loop,” in Proceedings of the IEEE/CVF international conference on computer vision , 2019, pp. 2252–2261

    2019

  5. [13]

    Score- guided diffusion for 3d human recovery,

    Anastasis Stathopoulos, Ligong Han, and Dimitris Metaxas, “Score- guided diffusion for 3d human recovery,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , 2024, pp. 906–915

    2024

  6. [14]

    Neural mocon: Neural motion control for physically plausible human motion capture,

    Buzhen Huang, Liang Pan, Yuan Yang, Jingyi Ju, and Yangang Wang, “Neural mocon: Neural motion control for physically plausible human motion capture,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , 2022, pp. 6417–6426

    2022

  7. [15]

    Closely interactive human reconstruction with proxemics and physics-guided adaption,

    Buzhen Huang, Chen Li, Chongyang Xu, Liang Pan, Yangang Wang, and Gim Hee Lee, “Closely interactive human reconstruction with proxemics and physics-guided adaption,” in IEEE/CVF Conference on Computer Vision and Pattern Recognition , 2024

    2024

  8. [16]

    Generating diverse and natural 3d human motions from text,

    Chuan Guo, Shihao Zou, Xinxin Zuo, Sen Wang, Wei Ji, Xingyu Li, and Li Cheng, “Generating diverse and natural 3d human motions from text,” in CVPR, June 2022, pp. 5152–5161

    2022

  9. [17]

    TMR: Text-to- motion retrieval using contrastive 3D human motion synthesis,

    Mathis Petrovich, Michael J. Black, and G ¨ul Varol, “TMR: Text-to- motion retrieval using contrastive 3D human motion synthesis,” in International Conference on Computer Vision (ICCV) , 2023

    2023

  10. [18]

    Motiongpt: Human motion as a foreign language,

    Biao Jiang, Xin Chen, Wen Liu, Jingyi Yu, Gang Yu, and Tao Chen, “Motiongpt: Human motion as a foreign language,” Advances in Neural Information Processing Systems , vol. 36, 2024

    2024

  11. [19]

    Chatpose: Chatting about 3d human pose,

    Yao Feng, Jing Lin, Sai Kumar Dwivedi, Yu Sun, Priyanka Patel, and Michael J. Black, “Chatpose: Chatting about 3d human pose,” in CVPR, 2024

    2024

  12. [20]

    An image is worth 16x16 words: Transformers for image recognition at scale,

    Alexey Dosovitskiy, Lucas Beyer, Alexander Kolesnikov, Dirk Weis- senborn, Xiaohua Zhai, Thomas Unterthiner, Mostafa Dehghani, Matthias Minderer, Georg Heigold, Sylvain Gelly, Jakob Uszkoreit, and Neil Houlsby, “An image is worth 16x16 words: Transformers for image recognition...

    2021

  13. [21]

    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 Trans. Graph. , vol. 34, no. 6, oct 2015

    2015

  14. [22]

    On the continuity of rotation representations in neural networks,

    Yi Zhou, Connelly Barnes, Lu Jingwan, Yang Jimei, and Li Hao, “On the continuity of rotation representations in neural networks,” in The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) , June 2019

    2019

  15. [23]

    High-resolution image synthesis with latent diffu- sion models,

    Robin Rombach, Andreas Blattmann, Dominik Lorenz, Patrick Esser, and Bj ¨orn Ommer, “High-resolution image synthesis with latent diffu- sion models,” 2021

    2021

  16. [24]

    Cliff: Carrying location information in full frames into human pose and shape estimation,

    Zhihao Li, Jianzhuang Liu, Zhensong Zhang, Songcen Xu, and Youliang Yan, “Cliff: Carrying location information in full frames into human pose and shape estimation,” in ECCV, 2022

    2022

  17. [25]

    Learning transferable visual models from natural language supervision,

    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

    2021

  18. [26]

    Neural discrete represen- tation learning,

    Aaron Van Den Oord, Oriol Vinyals, et al., “Neural discrete represen- tation learning,” NIPS, vol. 30, 2017

    2017

  19. [27]

    Human3.6m: Large scale datasets and predictive methods for 3d human sensing in natural environments,

    Catalin Ionescu, Dragos Papava, Vlad Olaru, and Cristian Sminchisescu, “Human3.6m: Large scale datasets and predictive methods for 3d human sensing in natural environments,” IEEE Transactions on Pattern Analysis and Machine Intelligence , vol. 36, no. 7, pp. 1325–1339, 2014

    2014

  20. [28]

    Microsoft coco: Common objects in context,

    Tsung-Yi Lin, Michael Maire, Serge Belongie, James Hays, Pietro Perona, Deva Ramanan, Piotr Dollar, and Larry Zitnick, “Microsoft coco: Common objects in context,” in ECCV. September 2014, European Conference on Computer Vision

    2014

  21. [29]

    2d human pose estimation: New benchmark and state of the art analysis,

    Mykhaylo Andriluka, Leonid Pishchulin, Peter Gehler, and Bernt Schiele, “2d human pose estimation: New benchmark and state of the art analysis,” in 2014 IEEE Conference on Computer Vision and Pattern Recognition, 2014, pp. 3686–3693

    2014

  22. [30]

    Monocular 3d human pose estimation in the wild using improved cnn supervision,

    Dushyant Mehta, Helge Rhodin, Dan Casas, Pascal Fua, Oleksandr Sotnychenko, Weipeng Xu, and Christian Theobalt, “Monocular 3d human pose estimation in the wild using improved cnn supervision,” in 3D Vision (3DV), 2017 Fifth International Conference on . IEEE, 2017

    2017

  23. [31]

    Recovering accurate 3d human pose in the wild using imus and a moving camera,

    Timo von Marcard, Roberto Henschel, Michael Black, Bodo Rosenhahn, and Gerard Pons-Moll, “Recovering accurate 3d human pose in the wild using imus and a moving camera,” in European Conference on Computer Vision (ECCV), sep 2018

    2018

  24. [32]

    Attention is all you need,

    Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Ł ukasz Kaiser, and Illia Polosukhin, “Attention is all you need,” in Advances in Neural Information Processing Systems, I. Guyon, U. V on Luxburg, S. Bengio, H. Wallach, R. Fergus, S. Vish...

    2017

  25. [33]

    Probabilistic modeling for human mesh recovery,

    Nikos Kolotouros, Georgios Pavlakos, Dinesh Jayaraman, and Kostas Daniilidis, “Probabilistic modeling for human mesh recovery,” in Proceedings of the IEEE/CVF international conference on computer vision, 2021, pp. 11605–11614

    2021

  26. [34]

    End-to-end recovery of human shape and pose,

    Angjoo Kanazawa, Michael J Black, David W Jacobs, and Jitendra Malik, “End-to-end recovery of human shape and pose,” in Proceedings of the IEEE conference on computer vision and pattern recognition , 2018, pp. 7122–7131

    2018

  27. [35]

    Danet: Decompose-and-aggregate network for 3d human shape and pose estimation,

    Hongwen Zhang, Jie Cao, Guo Lu, Wanli Ouyang, and Zhenan Sun, “Danet: Decompose-and-aggregate network for 3d human shape and pose estimation,” in Proceedings of the 27th ACM International Conference on Multimedia , 2019, pp. 935–944

    2019

  28. [36]

    Pymaf: 3d human pose and shape regression with pyramidal mesh alignment feedback loop,

    Hongwen Zhang, Yating Tian, Xinchi Zhou, Wanli Ouyang, Yebin Liu, Limin Wang, and Zhenan Sun, “Pymaf: 3d human pose and shape regression with pyramidal mesh alignment feedback loop,” in Proceedings of the IEEE/CVF international conference on computer vision, 2021, pp. 11446–11456

    2021

  29. [37]

    Pare: Part attention regressor for 3d human body estimation,

    Muhammed Kocabas, Chun-Hao P Huang, Otmar Hilliges, and Michael J Black, “Pare: Part attention regressor for 3d human body estimation,” in Proceedings of the IEEE/CVF international conference on computer vision , 2021, pp. 11127–11137

    2021

  30. [38]

    Pymaf-x: Towards well-aligned full- body model regression from monocular images,

    Hongwen Zhang, Yating Tian, Yuxiang Zhang, Mengcheng Li, Liang An, Zhenan Sun, and Yebin Liu, “Pymaf-x: Towards well-aligned full- body model regression from monocular images,” IEEE Transactions on Pattern Analysis and Machine Intelligence , vol. 45, no. 10, pp. 12287– 12303, 2023

    2023

  31. [39]

    Gpt-4 technical report,

    Achiam Josh, Adler Steven, Agarwal Sandhini, Ahmad Lama, Akkaya Ilge, Leoni Florencia, Aleman, Almeida Diogo, Altenschmidt Janko, Altman Sam, and Anadkat et al. Shyamal, “Gpt-4 technical report,” 2024. 6 Adapting Human Mesh Recovery with Vision-Language Feedback Supplementary ...

    2024

Pith tools

Reviewed August 9, 2026 · model on record in the stance chip above.