Semi-Automatic Generation of Training Data for Neural Networks for 6D Pose Estimation and Robotic Grasping

Johannes Rauer, Mohamed Aburaia, Wilfried Wöber
2020 Zenodo  
Machine-learning-based approaches for pose estimation are trained using annotated groundtruth data – images showing the object and information of its pose. In this work an approach to semiautomatically generate 6D pose-annotated data, using a movable marker and an articulated robot, is presented. A neural network for pose estimation is trained using datasets varying in size and type. The evaluation shows that small datasets recorded in the target domain and supplemented with augmented images
more » ... d to more robust results than larger synthetic datasets. The results demonstrate that a mobile manipulator using the proposed pose-estimation system could be deployed in real-life logistics applications to increase the level of automation.
doi:10.5281/zenodo.4084908 fatcat:7fscsh6qwbah5iyvfj3a5tsqty