mlsanityData to 14 Oct 2026

Perspective Flow Aggregation for Data-Limited 6D Object Pose Estimation

arXiv 2203.09836 ↗First posted 18 Mar 2022cs.CV cs.ROcited by 32

Not on the map. Why?

Abstract (arXiv)

Most recent 6D object pose estimation methods, including unsupervised ones, require many real training images. Unfortunately, for some applications, such as those in space or deep under water, acquiring real images, even unannotated, is virtually impossible. In this paper, we propose a method that can be trained solely on synthetic images, or optionally using a few additional real ones. Given a rough pose estimate obtained from a first network, it uses a second network to predict a dense 2D correspondence field between the image rendered using the rough pose and the real image and infers the required pose correction. This approach is much less sensitive to the domain shift between synthetic and real images than state-of-the-art methods. It performs on par with methods that require annotated real images for training when not using any, and outperforms them considerably when using as few as twenty real images.

Key facts. Cited by 32 papers in the map (data through 14 Oct 2026). Builds on 39 of its 58 references that are in the map. First posted on arXiv 18 Mar 2022.

Builds on

39 of 58 references.

References · 39

  1. 2020RAFT: Recurrent All-Pairs Field Transforms for Optical Flowcited by 2,415
  2. 2020In Search of Lost Domain Generalizationcited by 1,054
  3. 2017PWC-Net: CNNs for Optical Flow Using Pyramid, Warping, and Cost Volumecited by 968
  4. 2016FlowNet 2.0: Evolution of Optical Flow Estimation with Deep Networkscited by 784
  5. 2017PoseCNN: A Convolutional Neural Network for 6D Object Pose Estimation in Cluttered Scenescited by 775
  6. 2021Domain Generalization with MixStylecited by 566
  7. 2017Neural 3D Mesh Renderercited by 437
  8. 2018PVNet: Pixel-wise Voting Network for 6DoF Pose Estimationcited by 385
  9. 2019DenseFusion: 6D Object Pose Estimation by Iterative Dense Fusioncited by 355
  10. 2021A Fourier-based Framework for Domain Generalizationcited by 290
  11. 2020CosyPose: Consistent multi-view multi-object 6D pose estimationcited by 287
  12. 2018DeepIM: Deep Iterative Matching for 6D Pose Estimationcited by 283
27 more
  1. 2018BOP: Benchmark for 6D Object Pose Estimationcited by 254
  2. 2017SSD-6D: Making RGB-based 3D detection and 6D pose estimation great againcited by 243
  3. 2017BB8: A Scalable, Accurate, Robust to Partial Occlusion Method for Predicting the 3D Poses of Challenging Objects without Using Depthcited by 225
  4. 2019Pix2Pose: Pixel-Wise Coordinate Regression of Objects for 6D Pose Estimationcited by 218
  5. 2017Real-Time Seamless Single Shot 6D Object Pose Predictioncited by 210
  6. 2020Unsupervised Intra-domain Adaptation for Semantic Segmentation through Self-Supervisioncited by 198
  7. 2019DPOD: 6D Pose Object Detector and Refinercited by 189
  8. 2021Learning to Diversify for Single Domain Generalizationcited by 170
  9. 2020EPOS: Estimating 6D Pose of Objects with Symmetriescited by 148
  10. 2021GDRNPP: A Geometry-guided and Fully Learning-based Object Pose Estimatorcited by 148
  11. 2020Unsupervised Domain Adaptation via Structurally Regularized Deep Clusteringcited by 140
  12. 2020Gradually Vanishing Bridge for Adversarial Domain Adaptationcited by 130
  13. 2020HybridPose: 6D Object Pose Estimation under Hybrid Representationscited by 107
  14. 2018Segmentation-driven 6D Object Pose Estimationcited by 104
  15. 2019Single-Stage 6D Object Pose Estimationcited by 88
  16. 2021SO-Pose: Exploiting Self-Occlusion for Direct 6D Pose Estimationcited by 88
  17. 2015EpicFlow: Edge-Preserving Interpolation of Correspondences for Optical Flowcited by 80
  18. 2018Making Deep Heatmaps Robust to Partial Occlusions for 3D Object Pose Estimationcited by 74
  19. 2019Photorealistic Image Synthesis for Object Instance Detectioncited by 66
  20. 2019Satellite Pose Estimation Challenge: Dataset, Competition Design and Resultscited by 62
  21. 2015Learning Descriptors for Object Recognition and 3D Pose Estimationcited by 60
  22. 2018Deep Model-Based 6D Pose Refinement in RGBcited by 59
  23. 2019Multi-path Learning for Object Pose Estimation Across Domainscited by 52
  24. 2021Wide-Depth-Range 6D Object Pose Estimation in Spacecited by 45
  25. 2015Learning Analysis-by-Synthesis for 6D Pose Estimation in RGB-D Imagescited by 29
  26. 2017Feature Mapping for Learning Fast and Accurate 3D Pose Inference from Synthetic Imagescited by 15
  27. 2020DeepURL: Deep Pose Estimation Framework for Underwater Relative Localizationcited by 9

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2022: 1; 2023: 10; 2024: 12; 2025: 7; 2026: 2.

32 papers.

Sort:

By area:

  1. 2023BOP Challenge 2022 on Detection, Segmentation and Pose Estimation of Specific Rigid Objectscited by 70
  2. 2024BOP Challenge 2023 on Detection, Segmentation and Pose Estimation of Seen and Unseen Rigid Objectscited by 60
  3. 2024Deep Learning-Based Object Pose Estimation: A Comprehensive Surveycited by 40
  4. 2024GenFlow: Generalizable Recurrent Flow for 6D Pose Refinement of Novel Objectscited by 33
  5. 2023Challenges for Monocular 6D Object Pose Estimation in Roboticscited by 21
  6. 2023Shape-Constraint Recurrent Flow for 6D Object Pose Estimationcited by 17
  7. 2023HiPose: Hierarchical Binary Surface Encoding and Correspondence Pruning for RGB-D 6DoF Object Pose Estimationcited by 17
  8. 2023CheckerPose: Progressive Dense Keypoint Localization for Object Pose Estimation with Graph Neural Networkcited by 15
  9. 2024LaPose: Laplacian Mixture Shape Modeling for RGB-Based Category-Level Object Pose Estimationcited by 14
  10. 2025Co-op: Correspondence-based Novel Object Pose Estimationcited by 14
  11. 2023Rigidity-Aware Detection for 6D Object Pose Estimationcited by 13
  12. 2024MRC-Net: 6-DoF Pose Estimation with MultiScale Residual Correlationcited by 11
20 more
  1. 2022COPE: End-to-end trainable Constant Runtime Object Pose Estimationcited by 10
  2. 2023Pseudo Flow Consistency for Self-Supervised 6D Object Pose Estimationcited by 9
  3. 2025UniHOPE: A Unified Approach for Hand-Only and Hand-Object Pose Estimationcited by 6
  4. 2023Pseudo-keypoint RKHS Learning for Self-supervised 6DoF Pose Estimationcited by 4
  5. 2024Free-Moving Object Reconstruction and Pose Estimation with Virtual Cameracited by 3
  6. 2024Synthetica: Large Scale Synthetic Data for Robot Perceptioncited by 3
  7. 2025HccePose(BF): Predicting Front & Back Surfaces to Construct Ultra-Dense 2D-3D Correspondences for Pose Estimationcited by 3
  8. 2023Open Challenges for Monocular Single-shot 6D Object Pose Estimationcited by 2
  9. 2024Shape-biased Texture Agnostic Representations for Improved Textureless and Metallic Object Detection and 6D Pose Estimationcited by 2
  10. 2024End-to-End Probabilistic Geometry-Guided Regression for 6DoF Object Pose Estimationcited by 2
  11. 2023PViT-6D: Overclocking Vision Transformers for 6D Pose Estimation with Confidence-Level Prediction and Pose Tokenscited by 1
  12. 2024Leveraging Positional Encoding for Robust Multi-Reference-Based Object 6D Pose Estimationcited by 1
  13. 2024FAFA: Frequency-Aware Flow-Aided Self-Supervision for Underwater Object Pose Estimationcited by 1
  14. 2025SCFlow2: Plug-and-Play Object Pose Refiner with Shape-Constraint Scene Flowcited by 1
  15. 2025RefPose: Leveraging Reference Geometric Correspondences for Accurate 6D Pose Estimation of Unseen Objectscited by 1
  16. 2026Generative 6D Pose Estimation via Conditional Flow Matchingcited by 1
  17. 2024FAST GDRNPP: Improving the Speed of State-of-the-Art 6D Object Pose Estimationcited by 0
  18. 2025DLTPose: 6DoF Pose Estimation From Accurate Dense Surface Point Estimatescited by 0
  19. 2025RAG-6DPose: Retrieval-Augmented 6D Pose Estimation via Leveraging CAD as Knowledge Basecited by 0
  20. 2026Foundational feature fusion for conditional flow matching in 6D pose estimationcited by 0