6D Object Pose Estimation Using a Particle Filter With Better Initialization
Estimation of 6D object poses is a key issue in robotic grasping tasks. Recently, many high-performance learning-based methods have been introduced using robust deep learning techniques; however, applying these methods to real robot environments requires many ground truth 6D pose annotations for tra...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2023-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10034745/ |
| _version_ | 1811169727004803072 |
|---|---|
| author | Gijae Lee Jun-Sik Kim Seungryong Kim Kanggeon Kim |
| author_facet | Gijae Lee Jun-Sik Kim Seungryong Kim Kanggeon Kim |
| author_sort | Gijae Lee |
| collection | DOAJ |
| description | Estimation of 6D object poses is a key issue in robotic grasping tasks. Recently, many high-performance learning-based methods have been introduced using robust deep learning techniques; however, applying these methods to real robot environments requires many ground truth 6D pose annotations for training. To address this problem, we propose a template matching-based particle filter approach for 6D pose estimation; the proposed method does not require ground truth 6D poses. Although particle filter approaches can stochastically avoid local optima, they require adequate initial pose hypotheses for estimating an accurate 6D object pose. Therefore, we estimated an initial translation of the target object for accurately initializing a particle filter by developing a new deep network. Once the proposed centroid prediction network (CPN) is trained with a specific dataset, no additional training is required for new objects not in the dataset. We evaluated the performance of the CPN and the proposed 6D pose estimation method on benchmark datasets, which demonstrated that the CPN can predict the centroid for any object, including those not in the training data, and that our 6D pose estimation method outperforms existing methods for partially occluded objects. Finally, we tested a grasping task based on our proposed method using a real robot platform to demonstrate an application of our method to a downstream task. This experiment shows that our method can be applied to part assembly, bin picking, and object manipulation without large training datasets with 6D pose annotations. The code and models are available at: <uri>https://github.com/oorrppp2/Particle_filter_approach_6D_pose_estimation</uri>. |
| first_indexed | 2024-04-10T16:46:42Z |
| format | Article |
| id | doaj.art-91c019be1a80436dbb615b90e14a6be4 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-04-10T16:46:42Z |
| publishDate | 2023-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-91c019be1a80436dbb615b90e14a6be42023-02-08T00:00:56ZengIEEEIEEE Access2169-35362023-01-0111114511146210.1109/ACCESS.2023.3241250100347456D Object Pose Estimation Using a Particle Filter With Better InitializationGijae Lee0https://orcid.org/0000-0002-8231-8439Jun-Sik Kim1Seungryong Kim2https://orcid.org/0000-0003-2927-6273Kanggeon Kim3https://orcid.org/0000-0001-9467-2323Artificial Intelligence and Robotics Institute, Korea Institute of Science and Technology, Seoul, South KoreaArtificial Intelligence and Robotics Institute, Korea Institute of Science and Technology, Seoul, South KoreaDepartment of Computer Science and Engineering, Korea University, Seoul, South KoreaArtificial Intelligence and Robotics Institute, Korea Institute of Science and Technology, Seoul, South KoreaEstimation of 6D object poses is a key issue in robotic grasping tasks. Recently, many high-performance learning-based methods have been introduced using robust deep learning techniques; however, applying these methods to real robot environments requires many ground truth 6D pose annotations for training. To address this problem, we propose a template matching-based particle filter approach for 6D pose estimation; the proposed method does not require ground truth 6D poses. Although particle filter approaches can stochastically avoid local optima, they require adequate initial pose hypotheses for estimating an accurate 6D object pose. Therefore, we estimated an initial translation of the target object for accurately initializing a particle filter by developing a new deep network. Once the proposed centroid prediction network (CPN) is trained with a specific dataset, no additional training is required for new objects not in the dataset. We evaluated the performance of the CPN and the proposed 6D pose estimation method on benchmark datasets, which demonstrated that the CPN can predict the centroid for any object, including those not in the training data, and that our 6D pose estimation method outperforms existing methods for partially occluded objects. Finally, we tested a grasping task based on our proposed method using a real robot platform to demonstrate an application of our method to a downstream task. This experiment shows that our method can be applied to part assembly, bin picking, and object manipulation without large training datasets with 6D pose annotations. The code and models are available at: <uri>https://github.com/oorrppp2/Particle_filter_approach_6D_pose_estimation</uri>.https://ieeexplore.ieee.org/document/10034745/6D pose estimationcentroid prediction networkparticle filterrobotic grasping |
| spellingShingle | Gijae Lee Jun-Sik Kim Seungryong Kim Kanggeon Kim 6D Object Pose Estimation Using a Particle Filter With Better Initialization IEEE Access 6D pose estimation centroid prediction network particle filter robotic grasping |
| title | 6D Object Pose Estimation Using a Particle Filter With Better Initialization |
| title_full | 6D Object Pose Estimation Using a Particle Filter With Better Initialization |
| title_fullStr | 6D Object Pose Estimation Using a Particle Filter With Better Initialization |
| title_full_unstemmed | 6D Object Pose Estimation Using a Particle Filter With Better Initialization |
| title_short | 6D Object Pose Estimation Using a Particle Filter With Better Initialization |
| title_sort | 6d object pose estimation using a particle filter with better initialization |
| topic | 6D pose estimation centroid prediction network particle filter robotic grasping |
| url | https://ieeexplore.ieee.org/document/10034745/ |
| work_keys_str_mv | AT gijaelee 6dobjectposeestimationusingaparticlefilterwithbetterinitialization AT junsikkim 6dobjectposeestimationusingaparticlefilterwithbetterinitialization AT seungryongkim 6dobjectposeestimationusingaparticlefilterwithbetterinitialization AT kanggeonkim 6dobjectposeestimationusingaparticlefilterwithbetterinitialization |