mlsanityData to 14 Oct 2026

Understanding Pose and Appearance Disentanglement in 3D Human Pose Estimation

arXiv 2309.11667 ↗First posted 20 Sep 2023cs.CVcited by 2

Not on the map. Why?

Abstract (arXiv)

As 3D human pose estimation can now be achieved with very high accuracy in the supervised learning scenario, tackling the case where 3D pose annotations are not available has received increasing attention. In particular, several methods have proposed to learn image representations in a self-supervised fashion so as to disentangle the appearance information from the pose one. The methods then only need a small amount of supervised data to train a pose regressor using the pose-related latent vector as input, as it should be free of appearance information. In this paper, we carry out in-depth analysis to understand to what degree the state-of-the-art disentangled representation learning methods truly separate the appearance information from the pose one. First, we study disentanglement from the perspective of the self-supervised network, via diverse image synthesis experiments. Second, we investigate disentanglement with respect to the 3D pose regressor following an adversarial attack perspective. Specifically, we design an adversarial strategy focusing on generating natural appearance changes of the subject, and against which we could expect a disentangled network to be robust. Altogether, our analyses show that disentanglement in the three state-of-the-art disentangled representation learning frameworks if far from complete, and that their pose codes contain significant appearance information. We believe that our approach provides a valuable testbed to evaluate the degree of disentanglement of pose from appearance in self-supervised 3D human pose estimation.

Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 31 of its 42 references that are in the map. First posted on arXiv 20 Sep 2023.

Builds on

31 of 42 references.

References · 31

  1. 2014Explaining and Harnessing Adversarial Examplescited by 7,364
  2. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,415
  3. 2017mixup: Beyond Empirical Risk Minimizationcited by 5,183
  4. 2013Intriguing properties of neural networkscited by 5,155
  5. 2016Towards Evaluating the Robustness of Neural Networkscited by 2,998
  6. 2019CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Featurescited by 2,884
  7. 2014Conditional Generative Adversarial Netscited by 2,162
  8. 2016Adversarial examples in the physical worldcited by 1,769
  9. 2015DeepFool: a simple and accurate method to fool deep neural networkscited by 1,318
  10. 2019AugMix: A Simple Data Processing Method to Improve Robustness and Uncertaintycited by 901
  11. 2015The Limitations of Deep Learning in Adversarial Settingscited by 790
  12. 2016Monocular 3D Human Pose Estimation In The Wild Using Improved CNN Supervisioncited by 494
19 more
  1. 2019Evading Defenses to Transferable Adversarial Examples by Translation-Invariant Attackscited by 442
  2. 2017A simple yet effective baseline for 3d human pose estimationcited by 401
  3. 2018Adversarial Logit Pairingcited by 224
  4. 2016Coarse-to-Fine Volumetric Prediction for Single-Image 3D Human Posecited by 221
  5. 2017Generating Natural Adversarial Examplescited by 183
  6. 2017Unsupervised Learning of Disentangled Representations from Videocited by 139
  7. 2018Constructing Unrestricted Adversarial Examples with Generative Modelscited by 135
  8. 2017VNect: Real-time 3D Human Pose Estimation with a Single RGB Cameracited by 129
  9. 2019Sparse and Imperceivable Adversarial Attackscited by 95
  10. 2019SemanticAdv: Generating Adversarial Examples via Attribute-conditional Image Editingcited by 93
  11. 2019Unrestricted Adversarial Examples via Semantic Manipulationcited by 84
  12. 2017Lifting from the Deep: Convolutional 3D Pose Estimation from a Single Imagecited by 73
  13. 2018Unsupervised Geometry-Aware Representation for 3D Human Pose Estimationcited by 73
  14. 2019Semantic Adversarial Attacks: Parametric Transformations That Fool Deep Classifierscited by 57
  15. 2019Theory and Evaluation Metrics for Learning Disentangled Representationscited by 57
  16. 2017Harvesting Multiple Views for Marker-less 3D Human Pose Annotationscited by 48
  17. 2017Deep Multitask Architecture for Integrated 2D and 3D Human Sensingcited by 21
  18. 2019Neural Scene Decomposition for Multi-Person Motion Capturecited by 21
  19. 2020Measuring the Biases and Effectiveness of Content-Style Disentanglementcited by 14

Built on by

Cited by year

Citing papers per year: 2023: 1; 2024: 0; 2025: 1.

2 papers.

  1. 2023TIC-TAC: A Framework for Improved Covariance Estimation in Deep Heteroscedastic Regressioncited by 4
  2. 2025Towards Self-Supervised Covariance Estimation in Deep Heteroscedastic Regressioncited by 2