Abstract
The main task of object pose estimation is to predict the 3D rotation and 3D translation of an object in the current scene relative to a fixed object in the world coordinates. The most commonly used algorithm in pose estimation is based on the object characteristics or keypoint information for matching. The accuracy of these algorithms in pose estimation depends on whether the object surface characteristics are apparent. To solve the problem mentioned above, we propose a pose estimation algorithm using multilevel keypoint aggregation in the point cloud. First, we use a deep learning convolutional neural network to predict the keypoint positions in the point cloud. Then we estimate multiple poses at different levels according to the keypoints predicted above. Finally, we aggregate multiple poses into the final pose according to the weight of each pose. Our experiments show that our method outperforms other approaches in two datasets, YCB-Video and LineMOD.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Data Availability
The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.
References
Tremblay J, To T, Sundaralingam B (2018) Deep object pose estimation for semantic robotic grasping of household objects. Robot Learn Conf
Wang Z, et al. (2020) Grasping pose estimation for SCARA robot based on deep learning of point cloud. Int J Adv Manuf Technol 108(4):1–15
Zhu M et al (2014) Single image 3D object detection and pose estimation for grasping. IEEE Int Conf Robot Autom
Yu J et al (2014) A vision-based robotic grasping system using deep learning for 3D object recognition and pose estimation. IEEE Int Conf Robot Biomim
Deng X et al (2019) Self-supervised 6D Object Pose Estimation for Robot Manipulation. IEEE Int Conf Robot Autom
Lin HY, Liang SC, Chen YK (2020) Robotic Grasping with Multi-View Image Acquisition and Model-Based Pose Estimation. IEEE Sens J 1(1):99
Drost B et al (2010) Model globally, match locally: efficient and robust 3D object recognition. IEEE Comput Vis Pattern Recog
Drost BH, Ulrich M (2011) Recognition and pose determination of 3D objects in 3D scenes using geometric point pair descriptors and the generalized Hough Transform. EP
Hinterstoisser S et al (2016) Going Further with Point Pair Features. Eur Conf Comput Vis
Sundermeyer M et al (2016) Implicit 3D Orientation Learning for 6D Object Detection from RGB Images. Eur Conf Comput Vis
Labbé Y et al (2020) CosyPose: consistent multi-view multi-object 6D pose estimation. Eur Conf Comput Vis
Vock R et al (2019) Fast template matching and pose estimation in 3D point clouds. Comput Graph
Ye C et al (2016) Fast Hierarchical Template Matching Strategy for Real-Time Pose Estimation of Texture-Less Objects. Int Conf Intell Robot Appl
Xiang Y et al (2018) PoseCNN: a Convolutional Neural Network for 6D Object Pose Estimation in Cluttered Scenes. RSS
Wang C et al (2019) DenseFusion: 6D Object Pose Estimation by Iterative Dense Fusion. Conf Comput Vis Pattern Recog (CVPR)
Qi CR et al (2017) PointNet: deep Learning on Point Sets for 3D Classification and Segmentation. Conf Comput Vis Pattern Recog (CVPR)
Qi CR et al (2018) Frustum PointNets for 3D Object Detection from RGB-D Data. Conf Comput Vis Pattern Recog (CVPR)
He Y et al (2020) PVN3D: a Deep Point-wise 3D Keypoints Voting Network for 6DoF Pose Estimation. Conf Comput Vis Pattern Recog (CVPR)
Mo et al (2022) ES6D: a Computation Efficient and Symmetry-Aware 6D Pose Regression Framework. Conf Comput Vis Pattern Recog (CVPR)
Jiang M et al (2022) Uni6D: a Unified CNN Framework without Projection Breakdown for 6D Pose Estimation
Gao G et al (2020) 6D Object Pose Regression via Supervised Learning on Point Clouds. IEEE Int Conf Robot Autom (ICRA)
Shi G et al (2021) Fast Uncertainty Quantification for Deep Object Pose Estimation. IEEE Int Conf Robot Autom (ICRA)
Wang Z et al (2021) Simulation and deep learning on point clouds for robot grasping. Assem Autom
Shugurov, et al. (2022) OSOP: a multi-stage one shot object pose estimation framework. Conf Comput Vis Pattern Recog (CVPR
He K et al (2017) Mask R-CNN. IEEE Int Conf Comput Vis (ICCV)
Liu S et al (2018) Path Aggregation Network for Instance Segmentation. Conf Comput Vis Pattern Recog (CVPR)
Zhao H et al (2016) Pyramid Scene Parsing Network. IEEE Comput Soc
Xu D, Anguelov D, Jain A (2018) PointFusion: deep Sensor Fusion for 3D Bounding Box Estimation. Conf Comput Vis Pattern Recog (CVPR)
Yu P, Rao Y, et al. (2019) P2gnet: pose-guided point cloud generating networks for 6-dof object pose estimation. arXiv:1912.09316
Tian M et al (2020) Robust 6d object pose estimation by learning rgb-d features. Int Conf Robot Autom (ICRA)
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Statements and declarations
This work was supported by the Liaoning BaiQianWan Talents Program under Grant 2021222. This work was supported by the Natural Science Foundation of Liaoning Province (1600411972243).
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Junying Jia and Xin Lu are contributed equally to this work.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Yang, H., Jia, J. & Lu, X. A multilevel object pose estimation algorithm based on point cloud keypoints. Appl Intell 53, 18508–18516 (2023). https://doi.org/10.1007/s10489-022-04411-5
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10489-022-04411-5