FASTDLO: Fast Deformable Linear Objects Instance Segmentation

Abstract

In this paper, an approach for fast and accurate segmentation of Deformable Linear Objects (DLOs) named <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">FASTDLO</i> is presented. A deep convolutional neural network is employed for background segmentation, generating a binary mask that isolates DLOs in the image. Thereafter, the obtained mask is processed with a skeletonization algorithm and the intersections between different DLOs are solved with a similarity-based network. Apart from the usual pixel-wise color-mapped image, <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">FASTDLO</i> also describes each DLO instance with a sequence of 2D coordinates, enabling the possibility of modeling the DLO instances with splines curves, for example. Synthetically generated data are exploited for the training of the data-driven methods, avoiding expensive collection and annotations of real data. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">FASTDLO</i> is experimentally compared against both a DLO-specific approach and general-purpose deep learning instance segmentation models, achieving better overall performances and a processing rate higher than 20 FPS.