mlsanityData to 7 Oct 2026

Sim-to-Real 6D Object Pose Estimation via Iterative Self-training for Robotic Bin Picking

arXiv 2204.07049 ↗First posted 14 Apr 2022cs.RO cs.AI cs.CVcited by 20

Not on the map. Why?

Abstract (arXiv)

In this paper, we propose an iterative self-training framework for sim-to-real 6D object pose estimation to facilitate cost-effective robotic grasping. Given a bin-picking scenario, we establish a photo-realistic simulator to synthesize abundant virtual data, and use this to train an initial pose estimation network. This network then takes the role of a teacher model, which generates pose predictions for unlabeled real data. With these predictions, we further design a comprehensive adaptive selection scheme to distinguish reliable results, and leverage them as pseudo labels to update a student model for pose estimation on real data. To continuously improve the quality of pseudo labels, we iterate the above steps by taking the trained student model as a new teacher and re-label real data using the refined teacher model. We evaluate our method on a public benchmark and our newly-released dataset, achieving an ADD(-S) improvement of 11.49% and 22.62% respectively. Our method is also able to improve robotic bin-picking success by 19.54%, demonstrating the potential of iterative sim-to-real solutions for robotic applications.

Key facts. Cited by 20 papers in the map (data through 7 Oct 2026). Builds on 47 of its 69 references that are in the map. First posted on arXiv 14 Apr 2022.

Builds on

47 of 69 references.

References · 47

  1. 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,029
  2. 2017Mask R-CNNcited by 7,259
  3. 2015Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learningcited by 4,416
  4. 2017Domain Randomization for Transferring Deep Neural Networks from Simulation to the Real Worldcited by 1,401
  5. 2019Self-training with Noisy Student improves ImageNet classificationcited by 1,304
  6. 2019Unsupervised Data Augmentation for Consistency Trainingcited by 1,057
  7. 2017PoseCNN: A Convolutional Neural Network for 6D Object Pose Estimation in Cluttered Scenescited by 768
  8. 2020Simple Copy-Paste is a Strong Data Augmentation Method for Instance Segmentationcited by 595
  9. 2019Normalized Object Coordinate Space for Category-Level 6D Object Pose and Size Estimationcited by 490
  10. 2019DenseFusion: 6D Object Pose Estimation by Iterative Dense Fusioncited by 352
  11. 2021In Defense of Pseudo-Labeling: An Uncertainty-Aware Pseudo-label Selection Framework for Semi-Supervised Learningcited by 301
  12. 2018BOP: Benchmark for 6D Object Pose Estimationcited by 254
35 more
  1. 2017SSD-6D: Making RGB-based 3D detection and 6D pose estimation great againcited by 243
  2. 2018Deep Object Pose Estimation for Semantic Robotic Grasping of Household Objectscited by 237
  3. 2019Billion-scale semi-supervised learning for image classificationcited by 232
  4. 2019Learning to Predict 3D Objects with an Interpolation-based Differentiable Renderercited by 221
  5. 2020Atlas: End-to-End 3D Scene Reconstruction from Posed Imagescited by 206
  6. 2019PVN3D: A Deep Point-wise 3D Keypoints Voting Network for 6DoF Pose Estimationcited by 204
  7. 2020BOP Challenge 2020 on 6D Object Localizationcited by 192
  8. 2020Visual Camera Re-Localization from RGB and RGB-D Images Using DSACcited by 176
  9. 2018Sim-to-Real via Sim-to-Sim: Data-efficient Robotic Grasping via Randomized-to-Canonical Adaptation Networkscited by 171
  10. 2019Implicit 3D Orientation Learning for 6D Object Detection from RGB Imagescited by 170
  11. 2020Shape Prior Deformation for Categorical 6D Object Pose and Size Estimationcited by 158
  12. 2017Using Simulation and Domain Adaptation to Improve Efficiency of Deep Robotic Graspingcited by 138
  13. 2017Learning From Noisy Large-Scale Datasets With Minimal Supervisioncited by 115
  14. 2019Self-supervised 6D Object Pose Estimation for Robot Manipulationcited by 100
  15. 2018Falling Things: A Synthetic Dataset for 3D Object Detection and Pose Estimationcited by 92
  16. 2021SO-Pose: Exploiting Self-Occlusion for Direct 6D Pose Estimationcited by 88
  17. 2020Self6D: Self-Supervised Monocular 6D Object Pose Estimationcited by 79
  18. 2020Learning multiview 3D point cloud registrationcited by 78
  19. 2021Category-Level 6D Object Pose Estimation via Cascaded Relation and Recurrent Reconstruction Networkscited by 69
  20. 2017Transferring End-to-End Visuomotor Control from Simulation to Real World for a Multi-Stage Taskcited by 66
  21. 2019Automatic adaptation of object detectors to new domains using self-trainingcited by 66
  22. 2018Deep Model-Based 6D Pose Refinement in RGBcited by 59
  23. 2021A Closer Look at Self-training for Zero-Label Semantic Segmentationcited by 57
  24. 2019Multi-path Learning for Object Pose Estimation Across Domainscited by 52
  25. 2020MoreFusion: Multi-object Reasoning for 6D Pose Estimation from Volumetric Fusioncited by 46
  26. 2020Weakly-supervised 3D Shape Completion in the Wildcited by 33
  27. 2020Improving Semantic Segmentation via Self-Trainingcited by 30
  28. 2019Large-scale 6D Object Pose Estimation Dataset for Industrial Bin-Pickingcited by 25
  29. 2020Robust 6D Object Pose Estimation by Learning RGB-D Featurescited by 25
  30. 2021ROBI: A Multi-View Dataset for Reflective Objects in Robotic Bin-Pickingcited by 22
  31. 2020Single Shot 6D Object Pose Estimationcited by 19
  32. 2020PyraPose: Feature Pyramids for Fast and Accurate Object Pose Estimation under Domain Shiftcited by 14
  33. 2022ReorientBot: Learning Object Reorientation for Specific-Posed Placementcited by 10
  34. 2017Rotational Subgroup Voting and Pose Clustering for Robust 3D Object Recognitioncited by 6
  35. 2022SafePicking: Learning Safe Object Extraction via Object-Level Mappingcited by 3

Built on by

Cited by year

Citing papers per year: 2022: 3; 2023: 5; 2024: 6; 2025: 6.

20 papers.

Sort:

By area:

  1. 2023RoboTAP: Tracking Arbitrary Points for Few-Shot Visual Imitationcited by 88
  2. 2024Deep Learning-Based Object Pose Estimation: A Comprehensive Surveycited by 39
  3. 2024A Survey of Embodied Learning for Object-Centric Robotic Manipulationcited by 34
  4. 2022TexPose: Neural Texture Learning for Self-Supervised 6D Object Pose Estimationcited by 23
  5. 2023Deep Fusion Transformer Network with Weighted Vector-Wise Keypoints Voting for Robust 6D Object Pose Estimationcited by 16
  6. 2025Diff9D: Diffusion-Based Domain-Generalized Category-Level 9-DoF Object Pose Estimationcited by 16
  7. 2022Certifiable 3D Object Pose Estimation: Foundations, Learning Models, and Self-Trainingcited by 12
  8. 2025Scalable Unseen Objects 6-DoF Absolute Pose Estimation with Robotic Integrationcited by 11
  9. 2023Pseudo Flow Consistency for Self-Supervised 6D Object Pose Estimationcited by 9
  10. 2024Learning Instance-Aware Correspondences for Robust Multi-Instance Point Cloud Registration in Cluttered Scenescited by 7
  11. 2023A Correct-and-Certify Approach to Self-Supervise Object Pose Estimators via Ensemble Self-Trainingcited by 6
  12. 2022Lessons from a Space Lab -- An Image Acquisition Perspectivecited by 3
8 more
  1. 2024Test-Time Certifiable Self-Supervision to Bridge the Sim2Real Gap in Event-Based Satellite Pose Estimationcited by 3
  2. 2025ZeroBP: Learning Position-Aware Correspondence for Zero-shot 6D Pose Estimation in Bin-Pickingcited by 2
  3. 2025Reasoning and Learning a Perceptual Metric for Self-Training of Reflective Objects in Bin-Picking with a Low-cost Cameracited by 2
  4. 2024FAFA: Frequency-Aware Flow-Aided Self-Supervision for Underwater Object Pose Estimationcited by 1
  5. 2023Novel Synthetic Data Tool for Data-Driven Cardboard Box Localizationcited by 0
  6. 2024Targeted Hard Sample Synthesis Based on Estimated Pose and Occlusion Error for Improved Object Pose Estimationcited by 0
  7. 2025GMatch: A Lightweight, Geometry-Constrained Keypoint Matcher for Zero-Shot 6DoF Pose Estimation in Robotic Grasp Taskscited by 0
  8. 2025ROPES: Robotic Pose Estimation via Score-Based Causal Representation Learningcited by 0