mlsanityData to 7 Oct 2026

Leveraging 2D Masked Reconstruction for Domain Adaptation of 3D Pose Estimation

arXiv 2501.08408 ↗First posted 14 Jan 2025cs.CV cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

RGB-based 3D pose estimation methods have been successful with the development of deep learning and the emergence of high-quality 3D pose datasets. However, most existing methods do not operate well for testing images whose distribution is far from that of training data. However, most existing methods do not operate well for testing images whose distribution is far from that of training data. This problem might be alleviated by involving diverse data during training, however it is non-trivial to collect such diverse data with corresponding labels (i.e. 3D pose). In this paper, we introduced an unsupervised domain adaptation framework for 3D pose estimation that utilizes the unlabeled data in addition to labeled data via masked image modeling (MIM) framework. Foreground-centric reconstruction and attention regularization are further proposed to increase the effectiveness of unlabeled data usage. Experiments are conducted on the various datasets in human and hand pose estimation tasks, especially using the cross-domain scenario. We demonstrated the effectiveness of ours by achieving the state-of-the-art accuracy on all datasets.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 48 of its 67 references that are in the map. First posted on arXiv 14 Jan 2025.

Builds on

48 of 67 references.

References · 48

  1. 2017Attention Is All You Needcited by 59,932
  2. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  3. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  4. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,604
  5. 2014Microsoft COCO: Common Objects in Contextcited by 17,498
  6. 2020End-to-End Object Detection with Transformerscited by 7,384
  7. 2021Masked Autoencoders Are Scalable Vision Learnerscited by 7,339
  8. 2021Emerging Properties in Self-Supervised Vision Transformerscited by 6,893
  9. 2016SGDR: Stochastic Gradient Descent with Warm Restartscited by 4,993
  10. 2015Domain-Adversarial Training of Neural Networkscited by 3,023
  11. 2017Accurate, Large Minibatch SGD: Training ImageNet in 1 Hourcited by 1,479
  12. 2019Deep High-Resolution Representation Learning for Visual Recognitioncited by 1,417
36 more
  1. 2015Learning Transferable Features with Deep Adaptation Networkscited by 1,323
  2. 2014Deep Domain Confusion: Maximizing for Domain Invariancecited by 597
  3. 2016Monocular 3D Human Pose Estimation In The Wild Using Improved CNN Supervisioncited by 493
  4. 2020End-to-End Human Pose and Mesh Reconstruction with Transformerscited by 417
  5. 2017Learning from Synthetic Humanscited by 358
  6. 2019FreiHAND: A Dataset for Markerless Capture of Hand Pose and Shape from Single RGB Imagescited by 271
  7. 20213D Human Pose Estimation with Spatial and Temporal Transformerscited by 271
  8. 2021Mesh Graphormercited by 264
  9. 2018Total Capture: A 3D Deformation Model for Tracking Faces, Hands, and Bodiescited by 239
  10. 2017Learning to Estimate 3D Hand Pose from Single RGB Imagescited by 234
  11. 2021AGORA: Avatars in Geography Optimized for Regression Analysiscited by 203
  12. 2017Single-Shot Multi-Person 3D Pose Estimation From Monocular RGBcited by 195
  13. 2017GANerated Hands for Real-time 3D Hand Tracking from Monocular RGBcited by 147
  14. 2020TransPose: Keypoint Localization via Transformercited by 126
  15. 2021Keypoint Transformer: Solving Joint Identification in Challenging Hands and Object Interactions for Accurate 3D Pose Estimationcited by 122
  16. 2019Pushing the Envelope for RGB-based Dense 3D Hand Pose Estimation via Neural Renderingcited by 115
  17. 2021MST: Masked Self-Supervised Transformer for Visual Representationcited by 114
  18. 2018Hand Pose Estimation via Latent 2.5D Heatmap Regressioncited by 111
  19. 2019Cross-Domain Adaptation for Animal Pose Estimationcited by 108
  20. 2017V2V-PoseNet: Voxel-to-Voxel Prediction Network for Accurate 3D Hand and Human Pose Estimation from a Single Depth Mapcited by 102
  21. 2021Pose Recognition with Cascade Transformerscited by 99
  22. 2019Sim2real transfer learning for 3D human pose estimation: motion to the rescuecited by 75
  23. 2019Learning from Synthetic Animalscited by 65
  24. 2021TFPose: Direct Human Pose Estimation with Transformerscited by 42
  25. 2020Predicting Camera Viewpoint Improves Cross-dataset Generalization for 3D Human Pose Estimationcited by 36
  26. 2020Knowledge as Priors: Cross-Modal Knowledge Generalization for Datasets without Superior Knowledgecited by 35
  27. 2020Inference Stage Optimization for Cross-scenario 3D Human Pose Estimationcited by 35
  28. 2021Regressive Domain Adaptation for Unsupervised Keypoint Detectioncited by 34
  29. 2021AdaptPose: Cross-Dataset Adaptation for 3D Human Pose Estimation by Learnable Motion Generationcited by 32
  30. 2022Uncertainty-Aware Adaptation for Self-Supervised 3D Human Pose Estimationcited by 25
  31. 2022A Unified Framework for Domain Adaptive Pose Estimationcited by 20
  32. 2023Global Adaptation meets Local Generalization: Unsupervised Domain Adaptation for 3D Human Pose Estimationcited by 19
  33. 2018Augmented Skeleton Space Transfer for Depth-based Hand Pose Estimationcited by 15
  34. 2022TempCLR: Temporal Alignment Representation with Contrastive Learningcited by 15
  35. 2022Domain Adaptive Hand Keypoint and Pixel Localization in the Wildcited by 14
  36. 2022Multi-Person 3D Pose and Shape Estimation via Inverse Kinematics and Refinementcited by 11

Built on by

Not cited yet.