Skip to Main content Skip to Navigation
Journal articles

Virtual Training for a Real Application: Accurate Object-Robot Relative Localization without Calibration

Abstract : Localizing an object accurately with respect to a robot is a key step for autonomous robotic manipulation. In this work, we propose to tackle this task knowing only 3D models of the robot and object in the particular case where the scene is viewed from uncalibrated cameras — a situation which would be typical in an uncontrolled environment, e.g., on a construction site. We demonstrate that this localization can be performed very accurately, with millimetric errors, without using a single real image for training, a strong advantage since acquiring representative training data is a long and expensive process. Our approach relies on a classification Convolutional Neural Network (CNN) trained using hundreds of thousands of synthetically rendered scenes with randomized parameters. To evaluate our approach quantitatively and make it comparable to alternative approaches, we build a new rich dataset of real robot images with accurately localized blocks.
Document type :
Journal articles
Complete list of metadata

Cited literature [42 references]  Display  Hide  Download

https://hal-enpc.archives-ouvertes.fr/hal-01815826
Contributor : Romain Boistel Connect in order to contact the contributor
Submitted on : Wednesday, July 4, 2018 - 4:34:45 PM
Last modification on : Saturday, January 15, 2022 - 3:47:58 AM
Long-term archiving on: : Monday, October 1, 2018 - 1:30:11 PM

File

IJCV-2018-Loing-et-al.pdf
Files produced by the author(s)

Identifiers

Citation

Vianney Loing, Renaud Marlet, Mathieu Aubry. Virtual Training for a Real Application: Accurate Object-Robot Relative Localization without Calibration. International Journal of Computer Vision, Springer Verlag, In press, ⟨10.1007/s11263-018-1102-6⟩. ⟨hal-01815826v2⟩

Share

Metrics

Record views

370

Files downloads

217