On The Fly Generated Data for Industrial Part Orientation Estimation with Deep Neural Networks - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2019

On The Fly Generated Data for Industrial Part Orientation Estimation with Deep Neural Networks

Résumé

In this paper a method is proposed for estimating the orientation of industrial parts. Classical 2d images of the part are used to train a deep neural network and infer the part pose by computing its quaternion representation. Another innovative point of this work is in the use of synthetic data which are generated on the fly during the network training from a textured cad model placed in a virtual scene. This way overcomes the difficulties to obtain pose ground truth images from real images. At the same time, using the cad model with several lighting conditions and material reflectances allows to anticipate challenging industrial situations. As a first step of the method, the part is separated from the background using a semantic segmentation network. Then a depth image of the part is produced employing an encoder-decoder network with skip connections. Finally, the input image is associated with its depth map to estimate the part orientation with a fully connected network using a geodesic loss function. This method can estimate the part pose from a rendered image with a 5° accuracy while inferring from real images gives visually convincing results suitable for any pose refinement processes.
Fichier principal
Vignette du fichier
paper.pdf (617.44 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-02342703 , version 1 (01-11-2019)

Identifiants

Citer

Julien Langlois, Harold Mouchère, Nicolas Normand, Christian Viard-Gaudin. On The Fly Generated Data for Industrial Part Orientation Estimation with Deep Neural Networks. Fifteenth International Conference on Quality Control by Artificial Vision, May 2019, Mulhouse, France. pp.80, ⟨10.1117/12.2521753⟩. ⟨hal-02342703⟩
48 Consultations
122 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More