Training An Embedded Object Detector For Industrial Settings Without Real Images

Julia Cohen (1) , Carlos F Crispim-Junior (1) , Jean-Marc Chiappa , Laure Tougne (1)

1 imagine - Extraction de Caractéristiques et Identification

In an industrial environment, object detection is a challenging task due to the absence of real images and real-time requirements for the object detector, usually embedded in a mobile device. Using 3D models, it is however possible to create a synthetic dataset to train a neural network, although the performance on real images is limited by the domain gap. In this paper, we study the performance of a Convolutional Neural Network (CNN) designed to detect objects in real-time: Single-Shot Detector (SSD) with a MobileNet backbone. We train SSD with synthetic images only, and apply extensive data augmentation to reduce the domain gap between synthetic and real images. On the T-LESS dataset, SSD performs better than Mask R-CNN trained on the same synthetic images, with MobileNet-V2 and MobileNet-V3 Large as backbone. Our results also show the huge improvement enabled by an adequate augmentation strategy.

Mots clés: Object detection; Synthetic dataset; Mobile applications

Domaines: Intelligence artificielle [cs.AI]

Julia Cohen, Carlos F Crispim-Junior, Jean-Marc Chiappa, Laure Tougne. Training An Embedded Object Detector For Industrial Settings Without Real Images. 2021 IEEE International Conference on Image Processing (ICIP), Sep 2021, Anchorage, France. pp.714-718, ⟨10.1109/ICIP42928.2021.9506574⟩. ⟨hal-03531483⟩