Showing 1–2 of 2 results for author: Ruprecht, P

Deep 2.5D Vehicle Classification with Sparse SfM Depth Prior for Automated Toll Systems

Authors: Georg Waltner, Michael Maurer, Thomas Holzmann, Patrick Ruprecht, Michael Opitz, Horst Possegger, Friedrich Fraundorfer, Horst Bischof

Abstract: Automated toll systems rely on proper classification of the passing vehicles. This is especially difficult when the images used for classification only cover parts of the vehicle. To obtain information about the whole vehicle. we reconstruct the vehicle as 3D object and exploit this additional information within a Convolutional Neural Network (CNN). However, when using deep networks for 3D object… ▽ More Automated toll systems rely on proper classification of the passing vehicles. This is especially difficult when the images used for classification only cover parts of the vehicle. To obtain information about the whole vehicle. we reconstruct the vehicle as 3D object and exploit this additional information within a Convolutional Neural Network (CNN). However, when using deep networks for 3D object classification, large amounts of dense 3D models are required for good accuracy, which are often neither available nor feasible to process due to memory requirements. Therefore, in our method we reproject the 3D object onto the image plane using the reconstructed points, lines or both. We utilize this sparse depth prior within an auxiliary network branch that acts as a regularizer during training. We show that this auxiliary regularizer helps to improve accuracy compared to 2D classification on a real-world dataset. Furthermore due to the design of the network, at test time only the 2D camera images are required for classification which enables the usage in portable computer vision systems. △ Less

Submitted 11 May, 2018; v1 submitted 9 May, 2018; originally announced May 2018.

Comments: Submitted to the IEEE International Conference on Intelligent Transportation Systems 2018 (ITSC), 6 pages, 4 figures; changed format in compliance with adapted IEEE template

Collective excitations of atomic Bose-Einstein condensates

Authors: Mark Edwards, P. A. Ruprecht, K. Burnett, R. J. Dodd, Charles W. Clark

Abstract: We apply linear-response analysis of the Gross-Pitaevskii equation to obtain the excitation frequencies of a Bose-Einstein condensate confined in a time-averaged orbiting potential trap. Our calculated values are in excellent agreement with those observed in a recent experiment. We apply linear-response analysis of the Gross-Pitaevskii equation to obtain the excitation frequencies of a Bose-Einstein condensate confined in a time-averaged orbiting potential trap. Our calculated values are in excellent agreement with those observed in a recent experiment. △ Less

Submitted 28 May, 1996; originally announced May 1996.

Comments: 11 pages, 2 Postscript figures, uses psbox.tex for automatic figure inclusion. More info at http://amo.phy.gasou.edu/bec.html