“…The initial pictures used for the data generation can be taken with a studio setup and a rotation table for high resolution pictures but also in a minimal setup configuration. We tested our dataset generation with high quality pictures taken from [18] for one object, while for some other objects we used pictures taken with a standard smartphone on a basic black clothe. No significant difference was visible in the resulting performance of the model on our test cases.…”
Section: ) Synthetic Dataset Generation and Trainingmentioning
Fig. 1: Detection and pose estimation of two fallen objects on the ground. 3D models of the objects are placed on their estimated poses to compare the resulting RGB and depth images with the ones captured by the robot's camera.
“…The initial pictures used for the data generation can be taken with a studio setup and a rotation table for high resolution pictures but also in a minimal setup configuration. We tested our dataset generation with high quality pictures taken from [18] for one object, while for some other objects we used pictures taken with a standard smartphone on a basic black clothe. No significant difference was visible in the resulting performance of the model on our test cases.…”
Section: ) Synthetic Dataset Generation and Trainingmentioning
Fig. 1: Detection and pose estimation of two fallen objects on the ground. 3D models of the objects are placed on their estimated poses to compare the resulting RGB and depth images with the ones captured by the robot's camera.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.