6D位姿估计论文及代码汇总，持续更新~~~

左手的ㄟ右手 2022-12-03 09:10 182阅读 0赞

**转载请注明作者和出处：[http://blog.csdn.net/john\_bh/][http_blog.csdn.net_john_bh]**

**6D位姿估计 论文及代码汇总，持续更新中~~**

### 文章目录 ###

*   *  Arxiv
     *  ICCV 2021
     *  CVPR 2021
     *  WACV 2021
     *  ACCV 2020
     *  ECCV 2020
     *  CVPR 2020
     *  ICRA 2020
     *  IROS 2020
     *  WACV 2020
     *  CVPR 2019
     *  ICCV 2019
     *  NuerIPS 2019
     *  ICRA 2019
     *  ISRR 2019
     *  RSS 2019
     *  CoRL 2019
     *  CVPR 2018
     *  ECCV 2018
     *  RSS 2018
     *  CoRL 2018
     *  ICCV 2017
     *  CVPR 2017
     *  ICRA 2017
     *  CVPR 2016
     *  ECCV 2016
     *  ICRA 2016
     *  ICCV 2015
     *  ECCV 2014
     *  ACCV 2012

## Arxiv ##

1.  PAM:Point-wise Attention Module for 6D Object Pose Estimation — [arxiv paper][]
2.  PERCH 2.0 : Fast and Accurate GPU-based Perception via Search for Object Pose Estimation — [arxiv paper][arxiv paper 1]

## ICCV 2021 ##

1.  StereOBJ-1M: Large-scale Stereo Image Dataset for 6D Object Pose Estimation — [arxiv paper][arxiv paper 2]
2.  [SO-Pose: Exploiting Self-Occlusion for Direct 6D Pose Estimation][SO-Pose_ Exploiting Self-Occlusion for Direct 6D Pose Estimation] — [arxiv paper][arxiv paper 3] — [code][]

## CVPR 2021 ##

1.  [GDR-Net: Geometry-Guided Direct Regression Network for Monocular 6D Object Pose Estimation][GDR-Net_ Geometry-Guided Direct Regression Network for Monocular 6D Object Pose Estimation] — [paper][] — [code][code 1]
2.  DSC-PoseNet: Learning 6DoF Object Pose Estimation via Dual-scale Consistency — [arxiv paper][arxiv paper 4]
3.  StablePose: Learning 6D Object Poses from Geometrically Stable Patches — [arxiv paper][arxiv paper 5]

## WACV 2021 ##

1.  [A Pose Proposal and Refinement Network for Better 6D Object Pose Estimation][] — [`Best Student Paper` paper][Best Student Paper_ paper]

## ACCV 2020 ##

1.  A Sparse Gaussian Approach to Region-Based 6DoF Object Tracking — `Best Paper`[paper][paper 1] — [code][code 2]

## ECCV 2020 ##

1.  CosyPose：Consistent multi-view multi-object 6D pose estimation — [arxiv paper][arxiv paper 6] — [project][] — [code][code 3]
2.  Self6D：Self-Supervised Monocular 6D Object Pose Estimation — `Oral` [arxiv paper][arxiv paper 7] — [code][code 4]
3.  NOL：Neural Object Learning for 6D Pose Estimation Using a Few Cluttered Images — `Spotlight` [arxiv paper][arxiv paper 8] — [code][code 5] — [数据集][Link 1]
4.  [Category Level Object Pose Estimation via Neural Analysis-by-Synthesis][] — [arxiv paper][arxiv paper 9] — [主页][Link 2]
5.  Shape Prior Deformation：Shape Prior Deformation for Categorical 6D Object Pose and Size Estimation — [arxiv paper][arxiv paper 10] — [code][code 6]
6.  Few-Shot Viewpoint：Few-Shot Object Detection and Viewpoint Estimation for Objects in the Wild — [arxiv paper][arxiv paper 11] — [code][code 7]
7.  Geometric Correspondence Fields: Learned Differentiable Rendering for 3D Pose Refinement in the Wild — [arxiv paper][arxiv paper 12]

## CVPR 2020 ##

1.  PVN3D: A Deep Point-wise 3D Keypoints Voting Network for 6DoF Pose Estimation — [arxiv paper][arxiv paper 13] — [CVPR2020 paper][] — [code][code 8]
2.  Learning Canonical Shape Space for Category-Level 6D Object Pose and Size Estimation — [paper][paper 2]
3.  Category-Level Articulated Object Pose Estimation — `Oral` [paper][paper 3] — [code][code 9]
4.  MoreFusion: Multi-object Reasoning for 6D Pose Estimation from Volumetric Fusion — [arxiv paper][arxiv paper 14] — [code][code 10]
5.  EPOS: Estimating 6D Pose of Objects with Symmetries — [project][project 1] — [paper][paper 4]
6.  G2L-Net: Global to Local Network for Real-time 6D Pose Estimation with Embedding Vector Features — [arxiv paper][arxiv paper 15] — [code][code 11]
7.  [Hybridpose: 6d object pose estimation under hybrid representations ][Hybridpose_ 6d object pose estimation under hybrid representations]— [paper][paper 5] — [code][code 12]
8.  Multi-Path：Multi-path Learning for Object Pose Estimation Across Domains — [arxiv paper][arxiv paper 16] — [code][code 13]
9.  [LatentFusion: End-to-End Differentiable Reconstruction and Rendering for Unseen Object Pose Estimation][LatentFusion_ End-to-End Differentiable Reconstruction and Rendering for Unseen Object Pose Estimation] — [arxiv paper][arxiv paper 17] — [code][code 14] — [project][project 2]
10. Single-Stage 6D Object Pose Estimation — [arxiv paper][arxiv paper 18] — [code][code 15]
11. BPnP：End-to-End Learnable Geometric Vision by Backpropagating PnP Optimization — [arxiv paper][arxiv paper 19] — [code][code 16]
12. KeyPose: Multi-view 3D Labeling and Keypoint Estimation for Transparent Objects — [arxiv paper][arxiv paper 20] — [code][code 17]
13. [PFRL: Pose-Free Reinforcement Learning for 6D Pose Estimation][PFRL_ Pose-Free Reinforcement Learning for 6D Pose Estimation] — [paper][paper 6]
14. Reconstruct Locally, Localize Globally: A Model Free Method for Object Pose Estimation — [paper][paper 7]
15. Learning deep network for detecting 3D object keypoints and 6D poses — [paper][paper 8]
16. SSV：Self-Supervised Viewpoint Learning From Image Collections ---- [arxiv paper][arxiv paper 21] — [code][code 18]
17. Cylindrical Convolutional Networks for Joint Object Detection and Viewpoint Estimation — [arxiv paper][arxiv paper 22]

## ICRA 2020 ##

1.  6-PACK: Category-level 6D Pose Tracker with Anchor-Based Keypoints — [arxiv paper][arxiv paper 23] — [project][project 3] — [code][code 19]
2.  RotationAnchor：Robust 6D Object Pose Estimation by Learning RGB-D Features — [arxiv paper][arxiv paper 24] — [code][code 20]
3.  Self-Supervised 6D：Self-supervised 6D Object Pose Estimation for Robot Manipulation — [arxiv paper][arxiv paper 25]
4.  DirectShape: Direct Photometric Alignment of Shape Priors for Visual Vehicle Pose and Shape Estimatio — [arxiv paper][arxiv paper 26]

## IROS 2020 ##

1.  [se(3)-TrackNet: Data-driven 6D Pose Tracking by Calibrating Image Residuals in Synthetic Domains][se_3_-TrackNet_ Data-driven 6D Pose Tracking by Calibrating Image Residuals in Synthetic Domains] — [arxiv paper][arxiv paper 27] — [code][code 21]

## WACV 2020 ##

1.  PointPoseNet: Point Pose Network for Robust 6D Object Pose Estimation — [paper][paper 9]
2.  SymGAN: Orientation Estimation without Annotation for Symmetric Objects — [paper][paper 10]

## CVPR 2019 ##

1.  DenseFusion: 6D Object Pose Estimation by Iterative Dense Fusion — [arxiv paper][arxiv paper 28] — [project][project 4] — [code][code 22]
2.  [NOCS：Normalized Object Coordinate Space for Category-Level 6D Object Pose and Size Estimation][NOCS_Normalized Object Coordinate Space for Category-Level 6D Object Pose and Size Estimation] — [arxiv paper][arxiv paper 29] — [project][project 5] — [code][code 23]
3.  PVNet: Pixel-wise Voting Network for 6DoF Pose Estimation — [paper][paper 11] — [project][project 6] — [code][code 24]
4.  Segmentation-driven 6D object pose estimation – [paper][paper 12] — [code][code 25]

## ICCV 2019 ##

1.  CDPN: Coordinates-Based Disentangled Pose Network for Real-Time RGB-Based 6-DoF Object Pose Estimation — [paper][paper 13]
2.  Pix2Pose: Pixel-Wise Coordinate Regression of Objects for 6D Pose Estimation — [paper][paper 14] — [code][code 26]
3.  DPOD: 6D pose object detector and refiner — [paper][paper 15]

## NuerIPS 2019 ##

**Neural Information Processing Systems (NuerIPS)**

1.  Learning to predict 3D objects with an interpolation-based differentiable renderer

## ICRA 2019 ##

1.  MTTM：Multi-Task Template Matching for Object Detection, Segmentation and Pose Estimation Using Depth Images

## ISRR 2019 ##

1.  kPAM: KeyPoint Affordances for Category-Level Robotic Manipulation — [paper][paper 16]

## RSS 2019 ##

**Robotics: Science and Systems (RSS)**

1.  PoseRBPF: A rao-blackwellized particle filter for 6D object pose tracking — [paper][paper 17] — [code][code 27]

## CoRL 2019 ##

1.  Scene-level pose estimation for multiple instances of densely packed objects

## CVPR 2018 ##

1.  YOLO-6D：Real-Time Seamless Single Shot 6D Object Pose Prediction — [paper][paper 18] — [code][code 28]
2.  3D-RCNN:Instance-level 3D object reconstruction via render-andcompare
3.  Multiview consistency as supervisory signal for learning shape and pose prediction

## ECCV 2018 ##

1.  Category-level 6D Object Pose Recovery in Depth Images — [paper][paper 19]
2.  A unified framework for multi-view multi-class object pose estimation — [paper][paper 20]
3.  [DeepIM: Deep iterative matching for 6D pose estimation][DeepIM_ Deep iterative matching for 6D pose estimation] — [paper][paper 21] — [code][code 29]
4.  Implicit 3D orientation learning for 6D object detection from RGB images — [paper][paper 22] — [code][code 30]
5.  End-to-end 6-DOF object pose estimation through differentiable rasterization
6.  Making deep heatmaps robust to partial occlusions for 3D object pose estimation

## RSS 2018 ##

1.  [PoseCNN: A convolutional neural network for 6d object pose estimation in cluttered scenes][PoseCNN_ A convolutional neural network for 6d object pose estimation in cluttered scenes] — [paper][paper 23] — [code][code 31]

## CoRL 2018 ##

**Conference on Robot Learning**

1.  Deep object pose estimation for semantic robotic grasping of household objects
2.

## ICCV 2017 ##

1.  SSD-6D: Making RGB-based 3D detection and 6D pose estimation great again — [paper][paper 24] — [code][code 32]
2.  Real-Time Monocular Pose Estimation of 3D Objects using Temporally Consistent Local Color Histograms — [paper][paper 25]
3.  BB8: A Scalable, Accurate, Robust to Partial Occlusion Method for Predicting the 3D Poses of Challenging Objects without Using Depth — [paper][paper 26]

## CVPR 2017 ##

1.  3D bounding box estimation using deep learning and geometry
2.  Global hypothesis generation for 6D object pose estimation
3.  SSD-6D: Making RGB-based 3D detection and 6D pose estimation great again — [paper][paper 24] — [code][code 32]

## ICRA 2017 ##

1.  6-DOF object pose from semantic keypoints
2.  Multiview self-supervised deep learning for 6D pose estimation in the amazon picking challenge

## CVPR 2016 ##

1.  Uncertainty-driven 6D pose estimation of objects and scenes from a single RGB image

## ECCV 2016 ##

1.  Deep learning of local RGB-D patches for 3D object detection and 6d pose estimation — [paper][paper 27]

## ICRA 2016 ##

1.  Depth-based object tracking using a robust gaussian filter

## ICCV 2015 ##

1.  Learning analysis-by-synthesis for 6D pose estimation in RGB-D images

## ECCV 2014 ##

1.  Learning 6D object pose estimation using 3D object coordinates

## ACCV 2012 ##

1.  Model based training, detection and pose estimation of texture-less 3D objects in heavily cluttered scenes

[http_blog.csdn.net_john_bh]: http://blog.csdn.net/john_bh/
[arxiv paper]: https://arxiv.org/abs/2008.05242
[arxiv paper 1]: https://arxiv.org/abs/2008.00326
[arxiv paper 2]: https://arxiv.org/pdf/2109.10115.pdf
[SO-Pose_ Exploiting Self-Occlusion for Direct 6D Pose Estimation]: https://blog.csdn.net/john_bh/article/details/120491906
[arxiv paper 3]: https://arxiv.org/pdf/2108.08367.pdf
[code]: https://github.com/shangbuhuan13/SO-Pose
[GDR-Net_ Geometry-Guided Direct Regression Network for Monocular 6D Object Pose Estimation]: https://blog.csdn.net/john_bh/article/details/120023000
[paper]: https://openaccess.thecvf.com/content/CVPR2021/papers/Wang_GDR-Net_Geometry-Guided_Direct_Regression_Network_for_Monocular_6D_Object_Pose_CVPR_2021_paper.pdf
[code 1]: https://git.io/GDR-Net
[arxiv paper 4]: https://arxiv.org/pdf/2104.03658.pdf
[arxiv paper 5]: https://arxiv.org/pdf/2102.09334.pdf
[A Pose Proposal and Refinement Network for Better 6D Object Pose Estimation]: https://blog.csdn.net/john_bh/article/details/118100272
[Best Student Paper_ paper]: https://openaccess.thecvf.com/content/WACV2021/papers/Trabelsi_A_Pose_Proposal_and_Refinement_Network_for_Better_6D_Object_WACV_2021_paper.pdf
[paper 1]: https://openaccess.thecvf.com/content/ACCV2020/papers/Stoiber_A_Sparse_Gaussian_Approach_to_Region-Based_6DoF_Object_Tracking_ACCV_2020_paper.pdf
[code 2]: https://github.com/DLR-RM/RBGT
[arxiv paper 6]: https://arxiv.org/abs/2008.08465
[project]: https://www.di.ens.fr/willow/research/cosypose/
[code 3]: https://github.com/ylabbe/cosypose
[arxiv paper 7]: https://arxiv.org/abs/2004.06468
[code 4]: https://github.com/THU-DA-6D-Pose-Group/Self6D-Diff-Renderer
[arxiv paper 8]: https://arxiv.org/abs/2005.03717
[code 5]: https://github.com/kirumang/NOL
[Link 1]: https://www.acin.tuwien.ac.at/en/vision-for-robotics/software-tools/smot/
[Category Level Object Pose Estimation via Neural Analysis-by-Synthesis]: https://blog.csdn.net/john_bh/article/details/109625404
[arxiv paper 9]: https://arxiv.org/abs/2008.08145
[Link 2]: https://ait.ethz.ch/projects/2020/neural-object-fitting/
[arxiv paper 10]: https://arxiv.org/abs/2007.08454
[code 6]: https://github.com/mentian/object-deformnet
[arxiv paper 11]: https://arxiv.org/abs/2007.12107
[code 7]: https://github.com/YoungXIAO13/FewShotViewpoint
[arxiv paper 12]: https://arxiv.org/abs/2007.08939
[arxiv paper 13]: https://arxiv.org/abs/1911.04231
[CVPR2020 paper]: https://openaccess.thecvf.com/content_CVPR_2020/papers/He_PVN3D_A_Deep_Point-Wise_3D_Keypoints_Voting_Network_for_6DoF_CVPR_2020_paper.pdf
[code 8]: https://github.com/ethnhe/PVN3D
[paper 2]: https://openaccess.thecvf.com/content_CVPR_2020/papers/Chen_Learning_Canonical_Shape_Space_for_Category-Level_6D_Object_Pose_and_CVPR_2020_paper.pdf
[paper 3]: https://openaccess.thecvf.com/content_CVPR_2020/papers/Li_Category-Level_Articulated_Object_Pose_Estimation_CVPR_2020_paper.pdf
[code 9]: https://github.com/dragonlong/articulated-pose
[arxiv paper 14]: https://arxiv.org/abs/2004.04336
[code 10]: https://github.com/wkentaro/morefusion
[project 1]: http://cmp.felk.cvut.cz/epos/
[paper 4]: https://arxiv.org/abs/2004.00605
[arxiv paper 15]: https://arxiv.org/abs/2003.11089
[code 11]: https://github.com/DC1991/G2L_Net
[Hybridpose_ 6d object pose estimation under hybrid representations]: https://blog.csdn.net/john_bh/article/details/103998704
[paper 5]: https://openaccess.thecvf.com/content_CVPR_2020/papers/Song_HybridPose_6D_Object_Pose_Estimation_Under_Hybrid_Representations_CVPR_2020_paper.pdf
[code 12]: https://github.com/chensong1995/HybridPose
[arxiv paper 16]: https://arxiv.org/abs/1908.00151
[code 13]: https://github.com/DLR-RM/AugmentedAutoencoder/tree/multipath
[LatentFusion_ End-to-End Differentiable Reconstruction and Rendering for Unseen Object Pose Estimation]: https://blog.csdn.net/john_bh/article/details/109680518
[arxiv paper 17]: https://arxiv.org/abs/1912.00416
[code 14]: https://github.com/NVlabs/latentfusion
[project 2]: https://keunhong.com/publications/latentfusion/
[arxiv paper 18]: https://arxiv.org/pdf/1911.08324.pdf
[code 15]: https://github.com/cvlab-epfl/single-stage-pose
[arxiv paper 19]: https://arxiv.org/abs/1909.06043
[code 16]: https://github.com/BoChenYS/BPnP
[arxiv paper 20]: https://arxiv.org/abs/1912.02805
[code 17]: https://sites.google.com/view/keypose
[PFRL_ Pose-Free Reinforcement Learning for 6D Pose Estimation]: https://blog.csdn.net/john_bh/article/details/118391723
[paper 6]: http://openaccess.thecvf.com/content_CVPR_2020/papers/Shao_PFRL_Pose-Free_Reinforcement_Learning_for_6D_Pose_Estimation_CVPR_2020_paper.pdf
[paper 7]: http://openaccess.thecvf.com/content_CVPR_2020/papers/Cai_Reconstruct_Locally_Localize_Globally_A_Model_Free_Method_for_Object_CVPR_2020_paper.pdf
[paper 8]: https://openaccess.thecvf.com/content_CVPR_2020/papers/Zhao_Learning_Deep_Network_for_Detecting_3D_Object_Keypoints_and_6D_CVPR_2020_paper.pdf
[arxiv paper 21]: https://arxiv.org/abs/2004.01793
[code 18]: https://github.com/NVlabs/SSV
[arxiv paper 22]: https://arxiv.org/abs/2003.11303
[arxiv paper 23]: https://arxiv.org/abs/1910.10750
[project 3]: https://sites.google.com/view/6packtracking
[code 19]: https://github.com/j96w/6-PACK
[arxiv paper 24]: https://arxiv.org/abs/2003.00188
[code 20]: https://github.com/mentian/object-posenet
[arxiv paper 25]: https://arxiv.org/abs/1909.10159
[arxiv paper 26]: https://arxiv.org/abs/1904.10097
[se_3_-TrackNet_ Data-driven 6D Pose Tracking by Calibrating Image Residuals in Synthetic Domains]: https://blog.csdn.net/john_bh/article/details/109782035
[arxiv paper 27]: https://arxiv.org/pdf/2007.13866.pdf
[code 21]: https://github.com/wenbowen123/iros20-6d-pose-tracking
[paper 9]: https://openaccess.thecvf.com/content_WACV_2020/papers/Chen_PonitPoseNet_Point_Pose_Network_for_Robust_6D_Object_Pose_Estimation_WACV_2020_paper.pdf
[paper 10]: https://openaccess.thecvf.com/content_WACV_2020/papers/Ammirato_SymGAN_Orientation_Estimation_without_Annotation_for_Symmetric_Objects_WACV_2020_paper.pdf
[arxiv paper 28]: https://arxiv.org/abs/1901.04780
[project 4]: https://sites.google.com/view/densefusion/
[code 22]: https://github.com/j96w/DenseFusion
[NOCS_Normalized Object Coordinate Space for Category-Level 6D Object Pose and Size Estimation]: https://blog.csdn.net/john_bh/article/details/108377903
[arxiv paper 29]: https://arxiv.org/abs/1901.02970
[project 5]: https://geometry.stanford.edu/projects/NOCS_CVPR2019/
[code 23]: https://github.com/hughw19/NOCS_CVPR2019
[paper 11]: https://arxiv.org/pdf/1812.11788.pdf
[project 6]: https://zju3dv.github.io/pvnet/
[code 24]: https://github.com/zju3dv/pvnet
[paper 12]: https://openaccess.thecvf.com/content_CVPR_2019/papers/Hu_Segmentation-Driven_6D_Object_Pose_Estimation_CVPR_2019_paper.pdf
[code 25]: https://github.com/cvlab-epfl/segmentation-driven-pose
[paper 13]: https://openaccess.thecvf.com/content_ICCV_2019/papers/Li_CDPN_Coordinates-Based_Disentangled_Pose_Network_for_Real-Time_RGB-Based_6-DoF_Object_ICCV_2019_paper.pdf
[paper 14]: https://arxiv.org/abs/1908.07433
[code 26]: https://github.com/kirumang/Pix2Pose
[paper 15]: https://openaccess.thecvf.com/content_ICCV_2019/papers/Zakharov_DPOD_6D_Pose_Object_Detector_and_Refiner_ICCV_2019_paper.pdf
[paper 16]: https://arxiv.org/abs/1903.06684
[paper 17]: http://rss2019.informatik.uni-freiburg.de/papers/0152_FI.pdf
[code 27]: https://github.com/NVlabs/PoseRBPF
[paper 18]: https://arxiv.org/abs/1711.08848
[code 28]: https://github.com/Microsoft/singleshotpose
[paper 19]: https://openaccess.thecvf.com/content_ECCVW_2018/papers/11129/Sahin_Category-level_6D_Object_Pose_Recovery_in_Depth_Images_ECCVW_2018_paper.pdf
[paper 20]: https://openaccess.thecvf.com/content_ECCV_2018/papers/Chi_Li_A_Unified_Framework_ECCV_2018_paper.pdf
[DeepIM_ Deep iterative matching for 6D pose estimation]: https://blog.csdn.net/john_bh/article/details/119415760
[paper 21]: https://openaccess.thecvf.com/content_ECCV_2018/papers/Yi_Li_DeepIM_Deep_Iterative_ECCV_2018_paper.pdf
[code 29]: https://github.com/liyi14/mx-DeepIM
[paper 22]: https://openaccess.thecvf.com/content_ECCV_2018/papers/Martin_Sundermeyer_Implicit_3D_Orientation_ECCV_2018_paper.pdf
[code 30]: https://github.com/DLR-RM/AugmentedAutoencoder
[PoseCNN_ A convolutional neural network for 6d object pose estimation in cluttered scenes]: https://blog.csdn.net/john_bh/article/details/112312401
[paper 23]: https://arxiv.org/abs/1711.00199
[code 31]: https://github.com/yuxng/PoseCNN
[paper 24]: https://arxiv.org/abs/1711.10006v1
[code 32]: https://github.com/wadimkehl/ssd-6d
[paper 25]: https://openaccess.thecvf.com/content_ICCV_2017/papers/Tjaden_Real-Time_Monocular_Pose_ICCV_2017_paper.pdf
[paper 26]: https://openaccess.thecvf.com/content_ICCV_2017/papers/Rad_BB8_A_Scalable_ICCV_2017_paper.pdf
[paper 27]: https://arxiv.org/abs/1607.06038