혼재된 환경에서의 효율적 로봇 파지를 위한 3차원 물체 인식 알고리즘 개발
Development of an Efficient 3D Object Recognition Algorithm for Robotic Grasping in Cluttered Environments
2
조회
0
다운로드
3
인용
0
참고문헌
키워드
초록
3D object detection pipelines often incorporate RGB-based object detection methods such as YOLO, which detects the object classes and bounding boxes from the RGB image. However, in complex environments where objects are heavily cluttered, bounding box approaches may show degraded performance due to the overlapping bounding boxes. Mask based methods such as Mask R-CNN can handle such situation better thanks to their detailed object masks, but they require much longer time for data preparation compared to bounding box-based approaches. In this paper, we present a 3D object recognition pipeline which uses either the YOLO or Mask R-CNN real-time object detection algorithm, K-nearest clustering algorithm, mask reduction algorithm and finally Principal Component Analysis (PCA) alg orithm to efficiently detect 3D poses of objects in a complex environment. Furthermore, we also present an improved YOLO based 3D object detection algorithm that uses a prioritized heightmap clustering algorithm to handle overlapping bounding boxes. The suggested algorithms have successfully been used at the Artificial-Intelligence Robot Challenge (ARC) 2021 competition with excellent results.
Abstract
3D object detection pipelines often incorporate RGB-based object detection methods such as YOLO, which detects the object classes and bounding boxes from the RGB image. However, in complex environments where objects are heavily cluttered, bounding box approaches may show degraded performance due to the overlapping bounding boxes. Mask based methods such as Mask R-CNN can handle such situation better thanks to their detailed object masks, but they require much longer time for data preparation compared to bounding box-based approaches. In this paper, we present a 3D object recognition pipeline which uses either the YOLO or Mask R-CNN real-time object detection algorithm, K-nearest clustering algorithm, mask reduction algorithm and finally Principal Component Analysis (PCA) alg orithm to efficiently detect 3D poses of objects in a complex environment. Furthermore, we also present an improved YOLO based 3D object detection algorithm that uses a prioritized heightmap clustering algorithm to handle overlapping bounding boxes. The suggested algorithms have successfully been used at the Artificial-Intelligence Robot Challenge (ARC) 2021 competition with excellent results.
참고문헌 (18)
- A. Bochkovskiy, "YOLOv4 : Optimal Speed and Accuracy of Object Detection", arXiv:2004.10934, ~ (2020)
- K. He, "Mask R-CNN", 2017 IEEE International Conference on Computer Vision (ICCV) (2017)
- Y. Yoo, "Multimodal Anomaly Detection based on Deep Auto-Encoder for Object Slip Perception of Mobile Manipulation Robots", 2021 IEEE International Conference on Robotics and Automation (ICRA) (2021)
- W. Kehl, "SSD-6D : Making RGB-Based 3D Detection and 6D Pose Estimation Great Again", arXiv:1711.10006, ~ (2017)
- T. -Y. Lin, "Microsoft COCO : Common Objects in Context", European Conference on Computer Vision, pp.740~755 (2014)
- D . Müller, "homer@UniKoblenz:Winning Team of the RoboCup Virtual @Home Open Platform League 2021", RoboCup 2021: Robot World Cup XXIV (2022)
- F. B. van der Burgh, "Tech United Eindhoven @Home2019 Champions Paper", RoboCup 2019: Robot World Cup XXIII (2019)
- Stephen G. McGill, "Team THOR's Entry in the DARPA Robotics Challenge Finals 2015", Journal of Field Robotics, 34(4), pp.775~801 (2016), Wiley[DOI]
- Jae-Bong Yi, "Unified Software Platform for Intelligent Home Service Robots", Applied Sciences, 10(17), pp.5874~ (2020), MDPI AG[DOI]
- S. Zakharov, "DPOD : 6D Pose Object Detector and Refiner", Computer Vision and Pattern Recognition (2019)
- H. Sun, "A 3D Convolutional Neural Network Towards Real-Time Amodal 3D Object Detection", 2018 IEEE, 2018, ~
- B. Tekin, "Real-time seamless single shot 6d object pose prediction", Computer Vision and Pattern Recognition (2018)
- Y. Tanaka, "Hibikino-Musashi@Home 2019 Team Description Paper", arXiv:2006.01233, ~ (2019)
- "2 019 - TDP - Tech United Eindhoven - RoboCup@Home - Domestic Standard Platform"[URL]
- R. Adams, "Seeded region growing", IEEE Transactions on Pattern Analysis and Machine Intelligence, 16(6), pp.641~647 (1994), Institute of Electrical and Electronics Engineers (IEEE)[DOI]
- Jianping Fan, "Automatic image segmentation by integrating color-edge extraction and seeded region growing", IEEE Transactions on Image Processing, 10(10), pp.1454~1466 (2001), Institute of Electrical and Electronics Engineers (IEEE)[DOI]
- Hervé Abdi, "Principal component analysis", Wiley Interdisciplinary Reviews: Computational Statistics, 2(4), pp.433~459 (2010), Wiley[DOI]
- Taewoong Kang, "High-Speed Autonomous Robotic Assembly Using In-Hand Manipulation and Re-Grasping", Applied Sciences, 11(1), pp.37~ (2020), MDPI AG[DOI]