Machine Vision for Embedded Devices: from Synthetic Object Detection to Pyramidal Stereo Matching

In this work we present an embedded and all-in-one system for machine vision in industrial settings. This system enhances the capabilities of an industrial robot providing vision and perception, e.g. deep learning based object detection and 3D reconstruction by mean of efficient and highly scalable stereo matching. To this purpose we implemented and tested innovative solutions for object detection based on synthetically trained deep networks and a novel approach for depth estimation that embeds traditional 3D stereo matching within a pyramidal framework in order to reduce the computation time. Both object detection and 3D stereo matching have been efficiently implemented on the embedded device. Results and performance of the implementations are given for publicly available datasets, in particular the T-Less dataset for textureless object detection, Kitti Stereo and Middlebury Stereo datasets for depth estimation.