Pith. sign in

REVIEW 5 major objections 5 minor 60 references

Cross-Domain Adaptation for Animal Pose Estimation

T0 review · 5 major / 5 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read This paper claims that animal pose estimation can be transferred to quadruped species never seen in training by combining human pose knowledge, a small labeled animal set, and adversarial cross-domain adaptation with self-paced…

desk verdict Worth engaging for the Animal-Pose dataset and a sensible domain-adaptation recipe, but the abstract and conclusion overstate what was actually measured. read the letter →

arxiv 1908.05806 v2 pith:JRQCDGWT submitted 2019-08-16 cs.CV

Jinkun Cao , Hongyang Tang , Hao-Shu Fang , Xiaoyong Shen , Cewu Lu , Yu-Wing Tai This is my paper · ORCID
classification cs.CV
keywords animalposeestimationcross-domainadaptationpseudo-labelsself-pacedlearningadversarialdomainkeypointalignmenttransferunseenspecies
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

Animal pose estimation has been held back by the lack of labeled pose data for most species. This paper tries to remove that bottleneck by transferring pose knowledge from humans, whose skeletons resemble four-footed mammals, and from a small labeled set of five species to entirely unlabeled species. The authors build a pose-labeled dataset of dogs, cats, horses, sheep, and cows in a 17-keypoint format aligned with human pose annotation, then combine it with the large human pose dataset and box-labeled animal images. Their two-stage scheme, WS-CDA followed by PPLO, reaches 50+ mAP on held-out animal classes without any pose labels for those classes, and 65.7 mAP when 4,000 labeled animal instances are available. If the transfer holds, animal pose estimation becomes feasible for many species without per-species keypoint labeling.

What carries the argument

The load-bearing machinery is adversarial domain confusion in feature space combined with confidence-filtered pseudo-labels. WS-CDA's feature extractor feeds a domain discriminator that tries to tell humans, labeled animals, and unlabeled target animals apart; the main pose estimator is trained against that discriminator, so the features it learns are shared rather than species-specific. A weight w2 > 1 on the animal pose loss prevents the huge human dataset from dominating. PPLO uses a self-paced threshold, starting at 0.9 and decaying, to admit pseudo-labels only when the current model is confident, and alternating training between source and pseudo-labeled target minibatches prevents the larger source from overwhelming the target. Together these pieces turn human joints into usable supervision for quadrupeds.

What would settle it

Measure per-joint mAP on a held-out species: the homology assumption predicts the model should succeed on all 17 corresponding keypoints. If accuracy is high only for eyes, nose, and ears and collapses for elbows, knees, or withers, then the cross-species keypoint correspondence, not the adaptation machinery, is the load-bearing part of the claim.

Watch

Extended reading notes

Core claim

On the paper's own terms, the discovery is that cross-domain adaptation can make pose estimators generalize across species and even from humans to quadrupeds. WS-CDA uses an adversarial domain discriminator to force the feature extractor to keep only domain-invariant features, while a reweighted pose loss keeps the small labeled animal set from being swamped by the much larger human set. PPLO then selects the model's own confident predictions on unlabeled target species as pseudo-labels, raising the confidence threshold in a self-paced way and alternating source and target training to avoid noise and volume imbalance. The key claim is the result: on each of the five mammal classes held out one at a time, the full method is the best among the compared frameworks and adaptation methods, with per-class mAP from 41.0 to 57.3 and an average above 50, all without pose labels for the held-out class. The paper further states that this constitutes human-level pose estimation accuracy on animal pose estimation.

Load-bearing premise

The entire transfer rests on the assumption that the 17-keypoint format puts anatomically equivalent joints on humans and every target quadruped, so human pose priors and cross-species alignment teach the model the same body landmarks.

Editorial extensions

If this is right

  • A model trained on humans plus four labeled mammal classes can estimate pose on a fifth, unlabeled mammal class at above 50 mAP, with per-class results of 42.3, 41.0, 54.7, 57.3, and 53.1 mAP when that class is withheld.
  • Adding the large human dataset improves animal pose accuracy over training on labeled animals alone, even though human-only training fails completely on animal images.
  • Explicit adversarial domain adaptation beats GAN-based style transfer for this task, because image translation can move joint locations while leaving old pose labels attached.
  • Introducing even 50 labeled instances per target species raises mAP by roughly 7 to 20 points depending on species, showing the fully unsupervised setting still leaves room for improvement.
  • The same architecture, with the 17-keypoint COCO alignment, is a reusable starting point for other keypoint transfer tasks where skeletons are anatomically similar.

Reading between the lines

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

  • The paper's seven bounding-box-only novel species provide a ready test of whether the transfer generalizes beyond the original five species; measuring mAP against a small manual keypoint labeling on those classes would settle it.
  • The bone-length-ratio analysis implies a quantitative predictor: species whose average bone proportions lie closer to the human/animal manifold should transfer better, so ranking species by those ratios could forecast per-species mAP without training.
  • The failure cases suggest appearance-specific confusions, such as spines, horns, clothing, and low contrast, rather than pose-geometry errors; a follow-up could add adversarial appearance augmentation or class-conditional feature normalization to target those failures.
  • The self-paced pseudo-label schedule resembles curriculum learning and should transfer to other dense prediction tasks where source and target domains are imbalanced but structurally linked, such as hand or fish keypoint estimation.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

5 major / 5 minor

Summary. The paper addresses animal pose estimation under limited labeled data. It introduces an Animal-Pose dataset of five quadruped classes (5,517 instances, 20 keypoints aligned to COCO format) and proposes a cross-domain adaptation framework combining (i) WS-CDA, an adversarial domain-confusion scheme trained on human pose data, a small labeled animal set, and box-labeled unlabeled animal images, and (ii) PPLO, a self-paced pseudo-label self-training procedure. Table 1 shows that adding human data and WS-CDA components improves mAP on a held-out animal test set; Table 2 reports leave-one-species-out results in which WS-CDA+PPLO outperforms several baselines and adapted methods; Appendix C gives a supervised upper bound. The authors claim '50+ mAP on unseen animal classes' and 'human-level pose estimation accuracy' in the Abstract and Conclusion.

Significance. If the quantitative claims held, the paper would make a useful contribution: a public dataset, a clearly described domain-adaptation recipe for keypoints, and evidence that human pose priors help quadruped pose estimation. The ablations are internally consistent, the dataset release is valuable, and the supervised upper-bound experiment in Appendix C is a thoughtful calibration of what is lost by unsupervised transfer. However, the headline claims are not supported by the reported numbers, and the evaluation currently lacks the robustness evidence needed to establish the method's advantages.

major comments (5)
  1. [Abstract; Section 6 Conclusion] The claim that the method 'can achieve 50+ mAP on unseen animal classes without any pose-labeled data for it' is contradicted by Table 2: per-class mAPs for WS-CDA+PPLO are 42.3 (cat), 41.0 (dog), 54.7 (sheep), 57.3 (cow), and 53.1 (horse), averaging 49.68, with only three of five classes above 50. The Conclusion's 'human-level pose estimation accuracy' is also unsupported because no human-accuracy baseline is defined or measured. Please report the average accurately, qualify the claim to the five quadruped classes, and either provide a human baseline or remove the phrase 'human-level accuracy.'
  2. [Section 5.3, Table 2] All comparisons are single runs without variance or significance testing. The margins over the strongest adaptation baselines are small (e.g., WS-CDA+PPLO vs dom confusion [49]: cat 42.3 vs 38.0, dog 41.0 vs 37.7, horse 53.1 vs 48.5). Without multiple seeds or a paired statistical test, the reported improvements may not be reproducible. Please add repeated runs with mean and standard deviation, or at least a sensitivity analysis over seeds, and state whether the reported numbers are the best or average of runs.
  3. [Section 5.3, Appendix A] The general claim of transfer to 'unseen animal classes' is only evaluated for the five domestic quadruped species that share the same keypoint template defined in Section 3.1. Appendix A lists seven more distant categories (otter, antelope, bear, chimpanzee, rhino, bobcat, hippopotamus) but provides only bounding boxes and qualitative examples, with no COCO-format keypoint evaluation. As a result, the external validity of the cross-species claim is limited to near-neighbor quadruped transfer. Please either add quantitative keypoint results for the Appendix A categories or restrict the claim to morphologically close quadruped species.
  4. [Section 4.2, Section 5.1] PPLO is a self-training loop in which the model's own confident predictions are used as labels, and the key hyperparameters (initial mu=0.9, the 0.01/10-epoch relaxation schedule, w2=10, and alpha=-1, beta=500) are reported without a validation-based selection procedure or sensitivity analysis. Since the pseudo-label threshold directly controls the amount of self-generated supervision, it is important to show that the reported peak mAP is not an artifact of a threshold chosen after seeing the test distribution. Please provide ablation or sensitivity results for mu and the schedule, and clarify how the hyperparameters were selected.
  5. [Section 3.1, Figure 2] The proposed transfer assumes that the 17 COCO human keypoints are anatomically homologous to the selected animal keypoints. The paper does not validate this correspondence for the five quadruped species, nor for the Appendix A categories. If, for example, elbow and knee indices on a quadruped correspond to different joints than in humans, both the human prior and the pseudo-labels in PPLO would be trained against inconsistent supervision. Please provide evidence for the alignment (e.g., per-keypoint annotation agreement, a human-annotation consistency study, or a dedicated analysis of the Figure 2 bone-length ratios), or explicitly discuss this as a limitation.
minor comments (5)
  1. [Table 1 caption] The abbreviation 'RB' is used in the table header and caption but is not defined; please spell it out.
  2. [Equations (8)-(9)] The symbol Y-hat is used for the binary pseudo-label selection variable in Eq. (8), while Section 3.2 uses Y-hat for latent ground truth; please use distinct notation to avoid ambiguity.
  3. [Section 5.1] Please state explicitly whether the pose-unlabeled animal images from COCO2017 train are disjoint from the 1,117-instance test set used in Tables 1 and 2; this is needed to rule out test-set contamination through PPLO.
  4. [Figure 2] Bone-length proportions are shown without error bars or sample sizes; please add them or state clearly that the values are class means.
  5. [Algorithm 1] Line 7 says 'if confidence of K_I > mu'; for consistency with Eq. (9), write 'if C(K_I) > mu'.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the adaptation scheme and PPLO self-training are evaluated against held-out ground-truth keypoints, and the paper's self-citations are not load-bearing.

full rationale

The paper's claims are empirical rather than derivational, and the load-bearing steps do not reduce, by the paper's own equations, to their inputs. WS-CDA optimizes an adversarial domain-discrimination loss and supervised pose losses (Eqs. 3-5) on human and labeled animal data; PPLO (Eqs. 7-9 and Algorithm 1) then trains on target-domain pseudo-labels generated by the current model. While PPLO is self-referential in the sense that the target-domain training targets are the model's own confident predictions, the reported mAP values (Table 2) are computed on a held-out test set with ground-truth annotations, so the final evaluation is not statistically forced by the pseudo-labels. The paper does not conceal that pseudo-labels are model outputs (Sec. 4.2, Algorithm 1), and it explicitly guards against noise via a self-paced confidence threshold, which is a design choice rather than a disguised input. No fitted parameter is renamed as a prediction: hyperparameters such as alpha, beta, w1, w2, and the mu schedule are stated as fixed settings, not fitted to the test classes. The self-citations ([13], [14]) appear as related work and baseline framework choices; they are not invoked as uniqueness theorems or as the sole justification for the central transfer claim. The abstract's '50+ mAP' overstates the Table 2 average (49.68), but that is an accuracy/validity issue, not circularity. Therefore no circular step meeting the required evidence standard can be exhibited.

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

The central claim rests on the homology assumption between human and quadruped skeletons, the accuracy of the newly built dataset, the reliability of confidence-based pseudo-label selection, and the efficacy of adversarial domain confusion. No new physical or conceptual entities are postulated; the Animal-Pose dataset is a new resource rather than an invented entity in the ledger sense. The free parameters are hand-chosen and directly affect the reported mAP.

free parameters (4)
  • w2 = 10
    Weight on the animal pose loss in Equation 4, intended to counter the human-data volume imbalance. Chosen by hand; no sensitivity analysis or validation split reported.
  • mu initial value and schedule = 0.9, decreasing by 0.01 every 10 epochs if a pseudo-label is updated
    Confidence threshold for accepting pseudo-labels in Equation 9 and Algorithm 1. Directly controls the noise/quantity trade-off of pseudo-labels; no principled criterion is given.
  • alpha and beta = -1 and 500
    Weights in the combined loss L_WS-CDA equals alpha times L_DDL plus beta times L_pose (Equation 5). Hand-set; adversarial balance is not studied.
  • w1 = 1
    Weight in the domain discrimination loss (Equation 3). Not varied; the relative importance of the human/animal versus target/non-target terms is arbitrary.
assumptions (4)
  • domain assumption The 17-keypoint COCO human pose format is anatomically alignable to a common set of body landmarks on quadruped mammals with consistent meaning across species.
    This is the basis for transferring human pose priors to animals and for the cross-species bone-proportion analysis. Stated in Section 3.1 and Figure 2. If incorrect, the entire transfer mechanism rests on mismatched supervision.
  • domain assumption The Animal-Pose dataset annotations, extended from existing Poselets keypoints with knees added by the authors, are accurate enough to serve as ground truth for evaluation.
    Section 3.1 claims only slight labor work is involved, but no annotation quality metric such as inter-annotator agreement is reported, and the dataset is used for both training and test.
  • domain assumption The model's confidence score is a reliable proxy for the correctness of a predicted pseudo-label.
    The entire PPLO self-paced selection mechanism in Section 4.2.2 (Equations 8 and 9) depends on a monotonic relationship between confidence and correctness, which is not validated.
  • domain assumption Adversarial domain confusion (Equation 3 with alpha times beta less than zero) forces the feature extractor to learn features that transfer to the pose estimation task across domains.
    Borrowed from prior domain adaptation work such as Ganin and Lempitsky and Tzeng et al.; the paper does not prove that confusing the discriminator aligns features in a way that improves keypoint localization, especially when pose labels are sparse.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Cross-Domain Adaptation for Animal Pose Estimation." pith.science (2026). https://pith.science/paper/JRQCDGWT

@misc{pith2026190805806,
  author       = {Pith},
  title        = {Pith review of: Cross-Domain Adaptation for Animal Pose Estimation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/JRQCDGWT}},
  note         = {Machine review of arXiv:1908.05806}
}
read the original abstract

In this paper, we are interested in pose estimation of animals. Animals usually exhibit a wide range of variations on poses and there is no available animal pose dataset for training and testing. To address this problem, we build an animal pose dataset to facilitate training and evaluation. Considering the heavy labor needed to label dataset and it is impossible to label data for all concerned animal species, we, therefore, proposed a novel cross-domain adaptation method to transform the animal pose knowledge from labeled animal classes to unlabeled animal classes. We use the modest animal pose dataset to adapt learned knowledge to multiple animals species. Moreover, humans also share skeleton similarities with some animals (especially four-footed mammals). Therefore, the easily available human pose dataset, which is of a much larger scale than our labeled animal dataset, provides important prior knowledge to boost up the performance on animal pose estimation. Experiments show that our proposed method leverages these pieces of prior knowledge well and achieves convincing results on animal pose estimation.

Figures

Figures reproduced from arXiv: 1908.05806 by the authors.

Figure 1
Figure 1. Some samples from the Animal-Pose dataset. [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. The length proportion of each defined “bones” for [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Pipelines in WS-CDA. Lines with color describe the flow of features along different paths. “DDL” indicates the [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (7 more)
Figure 4
Figure 4. Figure 4: The overall process of our proposed scheme with [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]
Figure 5
Figure 5. Figure 5: cross-domain adaptation results by our proposed scheme on unseen animals included in our built dataset. [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]
Figure 6
Figure 6. Figure 6: cross-domain adaptation results by our proposed scheme on unseen animals not included in our built dataset. [PITH_FULL_IMAGE:figures/full_fig_p008_6.png]
Figure 7
Figure 7. Figure 7: 18 ’bones’ defined based on COCO-format key [PITH_FULL_IMAGE:figures/full_fig_p011_7.png]
Figure 8
Figure 8. Figure 8: Samples of seven novel animal categories with bounding box provided in our dataset. [PITH_FULL_IMAGE:figures/full_fig_p012_8.png]
Figure 9
Figure 9. Figure 9: Upper images are estimated pose by model trained without WS-CDA. Lower ones are obtained after model being [PITH_FULL_IMAGE:figures/full_fig_p012_9.png]
Figure 10
Figure 10. Figure 10: Samples of failure cases generated by our proposed methods on unseen animal categories. [PITH_FULL_IMAGE:figures/full_fig_p012_10.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

60 extracted references · 55 canonical work pages

  1. [49]

    Simultaneous deep transfer across domains and tasks

    Eric Tzeng, Judy Hoffman, Trevor Darrell, and Kate Saenko. Simultaneous deep transfer across domains and tasks. In ICCV, pages 4068–4076, 2015

    work page 2015

  2. [1]

    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 CVPR, pages 3686–3693, 2014

    work page 2014

  3. [2]

    Dataset of keypoints and foreground an- notations for all categories of pascal 2011, Feb 2012

    Lubomir Bourdev. Dataset of keypoints and foreground an- notations for all categories of pascal 2011, Feb 2012

    work page 2011

  4. [3]

    Poselets: Body part detectors trained using 3d human pose annotations

    Lubomir Bourdev and Jitendra Malik. Poselets: Body part detectors trained using 3d human pose annotations. InICCV, pages 1365–1372, 2009

    work page 2009

  5. [4]

    Unsupervised pixel- level domain adaptation with generative adversarial net- works

    Konstantinos Bousmalis, Nathan Silberman, David Dohan, Dumitru Erhan, and Dilip Krishnan. Unsupervised pixel- level domain adaptation with generative adversarial net- works. In CVPR, volume 1, page 7, 2017

    work page 2017

  6. [5]

    Partial Adversarial Domain Adaptation

    Zhangjie Cao, Lijia Ma, Mingsheng Long, and Jianmin Wang. Partial adversarial domain adaptation. CoRR, abs/1808.04205, 2018

    work page Pith review arXiv 2018

  7. [6]

    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 CVPR, pages 7291–7299, 2017

    work page 2017

  8. [7]

    Multisource do- main adaptation and its application to early detection of fa- tigue

    Rita Chattopadhyay, Qian Sun, Wei Fan, Ian Davidson, Sethuraman Panchanathan, and Jieping Ye. Multisource do- main adaptation and its application to early detection of fa- tigue. ACM Transactions on Knowledge Discovery from Data (TKDD), 6:18, 2012

    work page 2012

Show all 60 references
  1. [8]

    Cascaded pyramid network for multi-person pose estimation

    Yilun Chen, Zhicheng Wang, Yuxiang Peng, Zhiqiang Zhang, Gang Yu, and Jian Sun. Cascaded pyramid network for multi-person pose estimation. In CVPR, pages 7103– 7112, 2018

    2018

  2. [9]

    Self adversarial training for human pose estimation

    Chia-Jung Chou, Jui-Ting Chien, and Hwann-Tzong Chen. Self adversarial training for human pose estimation. CoRR, abs/1707.02439, 2017

    2017 arXiv

  3. [10]

    Vision-based hand pose estimation: A review

    Ali Erol, George Bebis, Mircea Nicolescu, Richard D Boyle, and Xander Twombly. Vision-based hand pose estimation: A review. CVIU, 108:52–73, 2007

    2007

  4. [11]

    Everingham, L

    M. Everingham, L. Van Gool, C. K. I. Williams, J. Winn, and A. Zisserman. The PASCAL Visual Object Classes Challenge 2011 (VOC2011) Results. http://www.pascal- network.org/challenges/VOC/voc2011/workshop/index.html

    2011

  5. [12]

    Real time head pose estimation with random regression forests

    Gabriele Fanelli, Juergen Gall, and Luc Van Gool. Real time head pose estimation with random regression forests. In CVPR, pages 617–624, 2011

    2011

  6. [13]

    Weakly and semi supervised human body part parsing via pose-guided knowledge trans- fer

    Hao-Shu Fang, Guansong Lu, Xiaolin Fang, Jianwen Xie, Yu-Wing Tai, and Cewu Lu. Weakly and semi supervised human body part parsing via pose-guided knowledge trans- fer. In CVPR, pages 70–78. IEEE, 2018

    2018

  7. [14]

    RMPE: Regional multi-person pose estimation

    Hao-Shu Fang, Shuqin Xie, Yu-Wing Tai, and Cewu Lu. RMPE: Regional multi-person pose estimation. In ICCV, 2017

    2017

  8. [15]

    Unsupervised domain adaptation by backpropagation

    Yaroslav Ganin and Victor Lempitsky. Unsupervised domain adaptation by backpropagation. In ICML, pages 1180–1189, 2015

    2015

  9. [16]

    Multi-evidence fil- tering and fusion for multi-label classification, object detec- tion and semantic segmentation based on weakly supervised learning

    Weifeng Ge, Sibei Yang, and Yizhou Yu. Multi-evidence fil- tering and fusion for multi-label classification, object detec- tion and semantic segmentation based on weakly supervised learning. In CVPR, pages 1277–1286, 2018

    2018

  10. [17]

    Fine-grained recognition in the wild: A multi-task domain adaptation ap- proach

    Timnit Gebru, Judy Hoffman, and Li Fei-Fei. Fine-grained recognition in the wild: A multi-task domain adaptation ap- proach. In ICCV, pages 1358–1367, 2017

    2017

  11. [18]

    Using k-poselets for detecting people and lo- calizing their keypoints

    Georgia Gkioxari, Bharath Hariharan, Ross Girshick, and Ji- tendra Malik. Using k-poselets for detecting people and lo- calizing their keypoints. In CVPR, pages 3582–3589, 2014

    2014

  12. [19]

    Domain adaptation for large-scale sentiment classification: A cdeep learning approach

    Xavier Glorot, Antoine Bordes, and Yoshua Bengio. Domain adaptation for large-scale sentiment classification: A cdeep learning approach. In ICML, pages 513–520, 2011

    2011

  13. [20]

    Generative adversarial nets

    Ian Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courville, and Yoshua Bengio. Generative adversarial nets. In NIPS, pages 2672–2680, 2014

    2014

  14. [21]

    Do- main adaptation for object recognition: An unsupervised ap- proach

    Raghuraman Gopalan, Ruonan Li, and Rama Chellappa. Do- main adaptation for object recognition: An unsupervised ap- proach. In ICCV, pages 999–1006, 2011

    2011

  15. [22]

    Mask r-cnn

    Kaiming He, Georgia Gkioxari, Piotr Doll ´ar, and Ross Gir- shick. Mask r-cnn. In CVPR, pages 2961–2969, 2017

    2017

  16. [23]

    Deep residual learning for image recognition

    Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Deep residual learning for image recognition. In CVPR, pages 770–778, 2016

    2016

  17. [24]

    Squeeze-and-excitation net- works

    Jie Hu, Li Shen, and Gang Sun. Squeeze-and-excitation net- works. 2018

    2018

  18. [25]

    Auggan: Cross domain adaptation with gan-based data augmentation

    Sheng-Wei Huang, Che-Tsung Lin, Shu-Ping Chen, Yen-Yi Wu, Po-Hao Hsu, and Shang-Hong Lai. Auggan: Cross domain adaptation with gan-based data augmentation. In ECCV, pages 718–731, 2018

    2018

  19. [26]

    Cross-domain weakly-supervised object de- tection through progressive domain adaptation

    Naoto Inoue, Ryosuke Furuta, Toshihiko Yamasaki, and Kiy- oharu Aizawa. Cross-domain weakly-supervised object de- tection through progressive domain adaptation. In CVPR, pages 5001–5009, 2018

    2018

  20. [27]

    Self-paced curriculum learning

    Lu Jiang, Deyu Meng, Qian Zhao, Shiguang Shan, and Alexander G Hauptmann. Self-paced curriculum learning. In AAAI, pages 2694–2700, 2015

    2015

  21. [28]

    Deep adversarial attention alignment for unsupervised domain adaptation: the benefit of target expectation maximization

    Guoliang Kang, Liang Zheng, Yan Yan, and Yi Yang. Deep adversarial attention alignment for unsupervised domain adaptation: the benefit of target expectation maximization. In ECCV, pages 401–416, 2018

    2018

  22. [29]

    Hand pose estimation and hand shape classification using multi-layered randomized decision forests

    Cem Keskin, Furkan Kırac ¸, Yunus Emre Kara, and Lale Akarun. Hand pose estimation and hand shape classification using multi-layered randomized decision forests. In ECCV, pages 852–863, 2012

    2012

  23. [30]

    Self- paced learning for latent variable models

    M Pawan Kumar, Benjamin Packer, and Daphne Koller. Self- paced learning for latent variable models. In NIPS, pages 1189–1197, 2010

    2010

  24. [31]

    Backpropagation applied to handwrit- ten zip code recognition

    Yann LeCun, Bernhard Boser, John S Denker, Donnie Henderson, Richard E Howard, Wayne Hubbard, and Lawrence D Jackel. Backpropagation applied to handwrit- ten zip code recognition. Neural Computation, 1:541–551, 1989

    1989

  25. [32]

    Ap- paratus and methods for head pose estimation and head ges- ture detection, Aug 2008

    Renxiang Li, Carl M Danielsen, and Cuneyt M Taskiran. Ap- paratus and methods for head pose estimation and head ges- ture detection, Aug 2008. US Patent 7,412,077

    2008

  26. [33]

    Gen- erative semantic manipulation with mask-contrasting gan

    Xiaodan Liang, Hao Zhang, Liang Lin, and Eric Xing. Gen- erative semantic manipulation with mask-contrasting gan. In ECCV, pages 558–573, 2018

    2018

  27. [34]

    https://github.com/cocodataset/cocoapi

    Tsung-Yi Lin. https://github.com/cocodataset/cocoapi

  28. [35]

    Microsoft coco: Common objects in context

    Tsung-Yi Lin, Michael Maire, Serge Belongie, James Hays, Pietro Perona, Deva Ramanan, Piotr Doll´ar, and C Lawrence Zitnick. Microsoft coco: Common objects in context. In ECCV, pages 740–755. Springer, 2014

    2014

  29. [36]

    Unsupervised domain adaptation with residual trans- fer networks

    Mingsheng Long, Han Zhu, Jianmin Wang, and Michael I Jordan. Unsupervised domain adaptation with residual trans- fer networks. In NIPS, pages 136–144, 2016

    2016

  30. [37]

    Instagan: Instance-aware image-to-image translation

    Sangwoo Mo, Minsu Cho, and Jinwoo Shin. Instagan: Instance-aware image-to-image translation. In ICLR, 2019

    2019

  31. [38]

    Head pose estimation in computer vision: A survey

    Erik Murphy-Chutorian and Mohan Manubhai Trivedi. Head pose estimation in computer vision: A survey. IEEE Trans. Pattern Anal. Mach. Intell., 31:607–626, 2009

    2009

  32. [39]

    O. M. Parkhi, A. Vedaldi, A. Zisserman, and C. V . Jawahar. Cats and dogs. In CVPR, 2012

    2012

  33. [40]

    Automatic differentiation in pytorch

    Adam Paszke, Sam Gross, Soumith Chintala, Gregory Chanan, Edward Yang, Zachary DeVito, Zeming Lin, Al- ban Desmaison, Luca Antiga, and Adam Lerer. Automatic differentiation in pytorch. In NIPS, 2017

    2017

  34. [41]

    Articulated peo- ple detection and pose estimation: Reshaping the future

    Leonid Pishchulin, Arjun Jain, Mykhaylo Andriluka, Thorsten Thorm ¨ahlen, and Bernt Schiele. Articulated peo- ple detection and pose estimation: Reshaping the future. In CVPR, pages 3178–3185, 2012

    2012

  35. [42]

    Interspecies knowledge transfer for facial keypoint detection

    Maheen Rashid, Xiuye Gu, and Yong Jae Lee. Interspecies knowledge transfer for facial keypoint detection. In CVPR, pages 6894–6903, 2017

    2017

  36. [43]

    An overview of gradient descent optimiza- tion algorithms

    Sebastian Ruder. An overview of gradient descent optimiza- tion algorithms. CoRR, abs/1609.04747, 2016

    2016 arXiv

  37. [44]

    Prototypical networks for few-shot learning

    Jake Snell, Kevin Swersky, and Richard Zemel. Prototypical networks for few-shot learning. In NIPS, pages 4077–4087, 2017

    2017

  38. [45]

    Zero-shot learning through cross-modal transfer

    Richard Socher, Milind Ganjoo, Christopher D Manning, and Andrew Ng. Zero-shot learning through cross-modal transfer. In NIPS, pages 935–943, 2013

    2013

  39. [46]

    Cross-modal deep variational hand pose estimation

    Adrian Spurr, Jie Song, Seonwook Park, and Otmar Hilliges. Cross-modal deep variational hand pose estimation. In CVPR, pages 89–98, 2018

    2018

  40. [47]

    Unsu- pervised learning of object landmarks by factorized spatial embeddings

    James Thewlis, Hakan Bilen, and Andrea Vedaldi. Unsu- pervised learning of object landmarks by factorized spatial embeddings. In ICCV, pages 5916–5925, 2017

    2017

  41. [48]

    Deeppose: Human pose estimation via deep neural networks

    Alexander Toshev and Christian Szegedy. Deeppose: Human pose estimation via deep neural networks. In CVPR, pages 1653–1660, 2014

    2014

  42. [50]

    Adversarial discriminative domain adaptation

    Eric Tzeng, Judy Hoffman, Kate Saenko, and Trevor Darrell. Adversarial discriminative domain adaptation. InCVPR, vol- ume 1, page 4, 2017

    2017

  43. [51]

    Understanding convolution for semantic segmentation

    Panqu Wang, Pengfei Chen, Ye Yuan, Ding Liu, Zehua Huang, Xiaodi Hou, and Garrison Cottrell. Understanding convolution for semantic segmentation. In WACV, pages 1451–1460, 2018

    2018

  44. [52]

    Human and sheep facial landmarks localisation by triplet interpo- lated features

    Heng Yang, Renqiao Zhang, and Peter Robinson. Human and sheep facial landmarks localisation by triplet interpo- lated features. In WACV, pages 1–8. IEEE, 2016

    2016

  45. [53]

    3d human pose estima- tion in the wild by adversarial learning

    Wei Yang, Wanli Ouyang, Xiaolong Wang, Jimmy Ren, Hongsheng Li, and Xiaogang Wang. 3d human pose estima- tion in the wild by adversarial learning. In CVPR, volume 1, 2018

    2018

  46. [54]

    Curricu- lum domain adaptation for semantic segmentation of urban scenes

    Yang Zhang, Philip David, and Boqing Gong. Curricu- lum domain adaptation for semantic segmentation of urban scenes. In ICCV, volume 2, page 6, 2017

    2017

  47. [55]

    Zero-shot learning via semantic similarity embedding

    Ziming Zhang and Venkatesh Saligrama. Zero-shot learning via semantic similarity embedding. In ICCV, pages 4166– 4174, 2015

    2015

  48. [56]

    Towards 3d human pose estimation in the wild: a weakly-supervised approach

    Xingyi Zhou, Qixing Huang, Xiao Sun, Xiangyang Xue, and Yichen Wei. Towards 3d human pose estimation in the wild: a weakly-supervised approach. In ICCV, 2017

    2017

  49. [57]

    Unsupervised domain adaptation for 3d key- point estimation via view consistency

    Xingyi Zhou, Arjun Karpur, Chuang Gan, Linjie Luo, and Qixing Huang. Unsupervised domain adaptation for 3d key- point estimation via view consistency. In ECCV, pages 137– 153, 2018

    2018

  50. [58]

    Unpaired image-to-image translation using cycle- consistent adversarial networks

    Jun-Yan Zhu, Taesung Park, Phillip Isola, and Alexei A Efros. Unpaired image-to-image translation using cycle- consistent adversarial networks. In ICCV, 2017

    2017

  51. [59]

    Unsupervised domain adaptation for semantic seg- mentation via class-balanced self-training

    Yang Zou, Zhiding Yu, BVK Vijaya Kumar, and Jinsong Wang. Unsupervised domain adaptation for semantic seg- mentation via class-balanced self-training. In ECCV, pages 289–305, 2018

    2018

  52. [60]

    3d menagerie: Modeling the 3d shape and pose of animals

    Silvia Zuffi, Angjoo Kanazawa, David W Jacobs, and Michael J Black. 3d menagerie: Modeling the 3d shape and pose of animals. In CVPR, pages 5524–5532, 2017. Appendix A. Proposed Dataset To bring convenience for the attempt to do domain adap- tation of animal pose estimation to ...

    2017

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.