RobotP: A Benchmark Dataset for 6D Object Pose Estimation
Deep learning has achieved great success on robotic vision tasks. However, when compared with other vision-based tasks, it is difficult to collect a representative and sufficiently large training set for six-dimensional (6D) object pose estimation, due to the inherent difficulty of data collection....
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-02-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/21/4/1299 |
| _version_ | 1797396865952514048 |
|---|---|
| author | Honglin Yuan Tim Hoogenkamp Remco C. Veltkamp |
| author_facet | Honglin Yuan Tim Hoogenkamp Remco C. Veltkamp |
| author_sort | Honglin Yuan |
| collection | DOAJ |
| description | Deep learning has achieved great success on robotic vision tasks. However, when compared with other vision-based tasks, it is difficult to collect a representative and sufficiently large training set for six-dimensional (6D) object pose estimation, due to the inherent difficulty of data collection. In this paper, we propose the RobotP dataset consisting of commonly used objects for benchmarking in 6D object pose estimation. To create the dataset, we apply a 3D reconstruction pipeline to produce high-quality depth images, ground truth poses, and 3D models for well-selected objects. Subsequently, based on the generated data, we produce object segmentation masks and two-dimensional (2D) bounding boxes automatically. To further enrich the data, we synthesize a large number of photo-realistic color-and-depth image pairs with ground truth 6D poses. Our dataset is freely distributed to research groups by the Shape Retrieval Challenge benchmark on 6D pose estimation. Based on our benchmark, different learning-based approaches are trained and tested by the unified dataset. The evaluation results indicate that there is considerable room for improvement in 6D object pose estimation, particularly for objects with dark colors, and photo-realistic images are helpful in increasing the performance of pose estimation algorithms. |
| first_indexed | 2024-03-09T00:57:40Z |
| format | Article |
| id | doaj.art-6877c2387d6d42c6acde3db38d254b86 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-09T00:57:40Z |
| publishDate | 2021-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-6877c2387d6d42c6acde3db38d254b862023-12-11T16:46:47ZengMDPI AGSensors1424-82202021-02-01214129910.3390/s21041299RobotP: A Benchmark Dataset for 6D Object Pose EstimationHonglin Yuan0Tim Hoogenkamp1Remco C. Veltkamp2Department of Information and Computing Sciences, Utrecht University, 3584 CC Utrecht, The NetherlandsDepartment of Information and Computing Sciences, Utrecht University, 3584 CC Utrecht, The NetherlandsDepartment of Information and Computing Sciences, Utrecht University, 3584 CC Utrecht, The NetherlandsDeep learning has achieved great success on robotic vision tasks. However, when compared with other vision-based tasks, it is difficult to collect a representative and sufficiently large training set for six-dimensional (6D) object pose estimation, due to the inherent difficulty of data collection. In this paper, we propose the RobotP dataset consisting of commonly used objects for benchmarking in 6D object pose estimation. To create the dataset, we apply a 3D reconstruction pipeline to produce high-quality depth images, ground truth poses, and 3D models for well-selected objects. Subsequently, based on the generated data, we produce object segmentation masks and two-dimensional (2D) bounding boxes automatically. To further enrich the data, we synthesize a large number of photo-realistic color-and-depth image pairs with ground truth 6D poses. Our dataset is freely distributed to research groups by the Shape Retrieval Challenge benchmark on 6D pose estimation. Based on our benchmark, different learning-based approaches are trained and tested by the unified dataset. The evaluation results indicate that there is considerable room for improvement in 6D object pose estimation, particularly for objects with dark colors, and photo-realistic images are helpful in increasing the performance of pose estimation algorithms.https://www.mdpi.com/1424-8220/21/4/1299benchmark dataset6D pose estimationsensors3D reconstruction |
| spellingShingle | Honglin Yuan Tim Hoogenkamp Remco C. Veltkamp RobotP: A Benchmark Dataset for 6D Object Pose Estimation Sensors benchmark dataset 6D pose estimation sensors 3D reconstruction |
| title | RobotP: A Benchmark Dataset for 6D Object Pose Estimation |
| title_full | RobotP: A Benchmark Dataset for 6D Object Pose Estimation |
| title_fullStr | RobotP: A Benchmark Dataset for 6D Object Pose Estimation |
| title_full_unstemmed | RobotP: A Benchmark Dataset for 6D Object Pose Estimation |
| title_short | RobotP: A Benchmark Dataset for 6D Object Pose Estimation |
| title_sort | robotp a benchmark dataset for 6d object pose estimation |
| topic | benchmark dataset 6D pose estimation sensors 3D reconstruction |
| url | https://www.mdpi.com/1424-8220/21/4/1299 |
| work_keys_str_mv | AT honglinyuan robotpabenchmarkdatasetfor6dobjectposeestimation AT timhoogenkamp robotpabenchmarkdatasetfor6dobjectposeestimation AT remcocveltkamp robotpabenchmarkdatasetfor6dobjectposeestimation |