Zobrazeno 1 - 10
of 12
pro vyhledávání: '"Jochen Abhau"'
Publikováno v:
IEEE Transactions on Intelligent Vehicles. 6:480-489
Object detectors are central to autonomous driving and are widely used in driver assistance systems. Object detectors are trained on a finite amount of data within a specific domain, hampering detection performance when applying object detectors to s
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::94c6395fd034be04b2e3e17e59036a43
https://doi.org/10.1109/tiv.2020.3039456
https://doi.org/10.1109/tiv.2020.3039456
Publikováno v:
Proceedings of the 2022 IEEE Intelligent Vehicles Symposium (IV)
Customization of a convolutional neural network (CNN) to a specific compute platform involves finding an optimal pareto state between computational complexity of the CNN and resulting throughput in operations per second on the compute platform. Howev
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::0395484bece43dce4f3efbd405573574
http://resolver.tudelft.nl/uuid:2f6fbb7a-b6ba-49b4-9740-c7306c994c48
http://resolver.tudelft.nl/uuid:2f6fbb7a-b6ba-49b4-9740-c7306c994c48
Publikováno v:
at - Automatisierungstechnik. 67:545-556
In autonomous driving, prediction tasks address complex spatio-temporal data. This article describes the examination of Recurrent Neural Networks (RNNs) for object trajectory prediction in the image space. The proposed methods enhance the performance
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::d1011b7af6c490e246cd496c65c45f21
https://doi.org/10.1515/auto-2018-0124
https://doi.org/10.1515/auto-2018-0124
Publikováno v:
AVSS
In deep learning applications large annotated datasets are considered necessary for application development and improved model performance. This work aims to investigate the validity of this assumption when enlarging a given dataset, by secondary dat
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::4236aeb39c9d861373a89278ef776244
https://doi.org/10.1109/avss.2019.8909858
https://doi.org/10.1109/avss.2019.8909858
Publikováno v:
Image and Vision Computing. 106:104079
Perception systems, to a large extent, rely on neural networks. Commonly, the training of neural networks uses a finite amount of data. The usual assumption is that an appropriate training dataset is available, which covers all relevant domains. This
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::986dd786aa31299792273fcf904d917c
https://doi.org/10.1016/j.imavis.2020.104079
https://doi.org/10.1016/j.imavis.2020.104079
Publikováno v:
Inverse Problems & Imaging. 7:1-25
In this paper we construct a shape space of medial ball representations from given shape training data using methods of Computational Geometry and Statistics. The ultimate goal is to employ the shape space as prior information in supervised segmentat
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::5e52256bc654757cbca2bfe8c7e28409
https://doi.org/10.3934/ipi.2013.7.1
https://doi.org/10.3934/ipi.2013.7.1
Autor:
Otmar Scherzer, Jochen Abhau
Publikováno v:
University of Vienna-u:cris
In this paper we propose an efficient algorithm for topology adaptation of evolving surface meshes in 3D. This system has two novel features: First, a spatial hashing technique is used to detect self-colliding triangles of the evolving mesh. Secondly
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::45b80355e107d2980aadc5a6c24e46b8
https://doi.org/10.1007/s11263-009-0282-5
https://doi.org/10.1007/s11263-009-0282-5
Publikováno v:
Lecture Notes in Computer Science ISBN: 9783642036408
EMMCVPR
University of Vienna-u:cris
EMMCVPR
University of Vienna-u:cris
In this paper we present a variational method for determining cartoon and texture components of the optical flow of a noisy image sequence. The method is realized by reformulating the optical flow problem first as a variational denoising problem for
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::cb25e3ba8dc9b3ab3852ec0936c84470
https://doi.org/10.1007/978-3-642-03641-5_10
https://doi.org/10.1007/978-3-642-03641-5_10
Publikováno v:
Geometry and Topology Monographs.
We report on the computation of the integral homology of the mapping class group of genus g surfaces with one boundary curve and m punctures, when 2g + m is smaller than 6. In particular, it includes the genus 2 case with no or one puncture.
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::1ce0f4a6fb0d2c97904e1646b080aecc
https://doi.org/10.2140/gtm.2008.14.1
https://doi.org/10.2140/gtm.2008.14.1
Autor:
Otmar Scherzer, Jochen Abhau
Publikováno v:
Medical Imaging: Image Processing
Active contour models have already been used succesfully for segmentation of organs from medical images in 3D. In implicit models, the contour is given as the isosurface of a scalar function, and therefore topology adaptations are handled naturally d
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::d3836130323f0973a5a017414c630fd1
https://doi.org/10.1117/12.770523
https://doi.org/10.1117/12.770523