[Zurück]

T. Heitzinger, M. Kampel:
"Highly Accurate Binary Image Segmentation for Cars";
in: "Proceedings of the Joint Austrian Computer Vision and Robotics Workshop 2020", 1; P. Roth et al. (Hrg.); herausgegeben von: Verlag der Technischen Universität Graz; Austrian Association for Pattern Recognition (ÖAGM/AAPR), Graz, 2020, ISBN: 978-3-85125-752-6, S. 116 - 121.

We study methods for the generation of highly accurate binary segmentation masks with application to images of cars. The goal is the automated separation of cars from their background. A fully convolutional network (FCN) based on the UNet architecture is trained on a private dataset consisting of over 7000 samples. The main contributions of the paper include a series of modification to common
loss functions as well as the introduction of a novel Gradient Loss that outperforms standard approaches. In a specialized postprocessing step the generated masks are further refined to better match the inherent curvature bias typically found in the outline of cars. In direct comparison to previous implementations our method reduces the segmentation error measured by the Jaccard index by over 65%.

http://dx.doi.org/10.3217/978-3-85125-752-6