Zobrazeno 1 - 10
of 15
pro vyhledávání: '"Manuel Lopez Antequera"'
Autor:
Maria Leyva-Vallina, Nicola Strisciuglio, Manuel Lopez Antequera, Radim Tylecek, Michael Blaich, Nicolai Petkov
Publikováno v:
IEEE Access, Vol 7, Pp 52277-52287 (2019)
Place recognition can be achieved by identifying whether a pair of images (a labeled reference image and a query image) depict the same place, regardless of appearance changes due to different viewpoints or lighting conditions. It is an important com
Externí odkaz:
https://doaj.org/article/9436b62c43194789bbec29d8646f46bd
Autor:
Samuel Rota Bulò, Manuel Lopez Antequera, Peter Kontschieder, Lorenzo Porzi, Andrea Simonelli
Publikováno v:
IEEE Transactions on Pattern Analysis and Machine Intelligence. 44:1219-1231
In this paper we introduce a method for multi-class, monocular 3D object detection from a single RGB image, which exploits a novel disentangling transformation and a novel, self-supervised confidence estimation method for predicted 3D bounding boxes.
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::51648437905be13d1a3ff2a59586ad13
https://doi.org/10.1109/tpami.2020.3025077
https://doi.org/10.1109/tpami.2020.3025077
Publikováno v:
Neural Computing and Applications, 32(24), 17957-17971. Springer
Neural Computing and Applications, 32(24), 17957-17971. SPRINGER LONDON LTD
Neural Computing and Applications, 32(24), 17957-17971. SPRINGER LONDON LTD
Convolutional neural networks (CNNs) lack robustness to test image corruptions that are not seen during training. In this paper, we propose a new layer for CNNs that increases their robustness to several types of corruptions of the input images. We c
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::65bac41c7d0a326d33be50c3ffe53341
https://doi.org/10.1007/s00521-020-04751-8
https://doi.org/10.1007/s00521-020-04751-8
Autor:
Nicolai Petkov, Radim Tylecek, Michael Blaich, Nicola Strisciuglio, Manuel Lopez Antequera, Maria Leyva-Vallina
Publikováno v:
IEEE Access, Vol 7, Pp 52277-52287 (2019)
IEEE Access, 7:8698240, 52277-52287. IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
IEEE Access, 7:8698240, 52277-52287. IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
Place recognition can be achieved by identifying whether a pair of images (a labeled reference image and a query image) depict the same place, regardless of appearance changes due to different viewpoints or lighting conditions. It is an important com
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::46a9c4945b3712efec5d4de4f114352c
https://doi.org/10.1109/ACCESS.2019.2910150
https://doi.org/10.1109/ACCESS.2019.2910150
Publikováno v:
IEEE Transactions on Robotics. 34:1623-1635
This paper presents a dense method for estimating planar motion with a laser scanner. Starting from a symmetric representation of geometric consistency between scans, we derive a precise range flow constraint and express the motion of the scan observ
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::5f5fbc46bd51aa14708c0d1d12d8dc34
https://doi.org/10.1109/tro.2018.2861911
https://doi.org/10.1109/tro.2018.2861911
Autor:
Manuel Lopez Antequera
This thesis explores several practical applications of computer vision using learning based techniques.We begin by exploring the task of single image camera calibration. That is, the prediction of a camera's orientation, zoom and distortion from sing
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::180418d098e486af05c7f0f7320aab07
https://doi.org/10.33612/diss.112968625
https://doi.org/10.33612/diss.112968625
Autor:
Samuel Rota Bulò, Markus Hofinger, Peter Kontschieder, Manuel Lopez Antequera, Yubin Kuang, Pau Gargallo
Publikováno v:
Computer Vision – ECCV 2020 ISBN: 9783030585358
ECCV (2)
ECCV (2)
Learning-based methods produce remarkable results on single image depth tasks when trained on well-established benchmarks, however, there is a large gap from these benchmarks to real-world performance that is usually obscured by the common practice o
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::4bfada2f5bf650b224c43610705646a0
https://doi.org/10.1007/978-3-030-58536-5_35
https://doi.org/10.1007/978-3-030-58536-5_35
Autor:
Søren Hauberg, Pau Gargallo, Manuel Lopez-Antequera, Yubin Kuang, Frederik Warburg, Javier Civera
Publikováno v:
Zaguán. Repositorio Digital de la Universidad de Zaragoza
instname
CVPR
Technical University of Denmark Orbit
Warburg, F R, Hauberg, S, Lopez-Antequera, M, Gargallo, P, Kuang, Y & Civera, J 2020, Mapillary Street-Level Sequences: A Dataset for Lifelong Place Recognition . in Proceedings of IEEE Conference on Computer Vision and Pattern Recognition 2020 . IEEE, 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Seattle, Washington, United States, 14/06/2020 .
instname
CVPR
Technical University of Denmark Orbit
Warburg, F R, Hauberg, S, Lopez-Antequera, M, Gargallo, P, Kuang, Y & Civera, J 2020, Mapillary Street-Level Sequences: A Dataset for Lifelong Place Recognition . in Proceedings of IEEE Conference on Computer Vision and Pattern Recognition 2020 . IEEE, 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Seattle, Washington, United States, 14/06/2020 .
Lifelong place recognition is an essential and challenging task in computer vision, with vast applications in robust localization and efficient large-scale 3D reconstruction. Progress is currently hindered by a lack of large, diverse, publicly availa
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::964613ce7ccf4612c35d67294ed145da
http://zaguan.unizar.es/record/106609
http://zaguan.unizar.es/record/106609
Publikováno v:
Pattern Recognition Letters, 92, 89-95. ELSEVIER SCIENCE BV
A convolutional neural network embedding to perform place recognition is introduced.A triplet similarity loss is chosen to allow for weakly supervised training.The network is trained with triplets of images presenting seasonal or other changes.The me
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::75f240913612a07e5c5cff6b92d4db6c
https://research.rug.nl/en/publications/4743d0ea-7b0c-49f5-a699-72335f3e0624
https://research.rug.nl/en/publications/4743d0ea-7b0c-49f5-a699-72335f3e0624
Autor:
Samuel Rota Bulò, Peter Kontschieder, Lorenzo Porzi, Andrea Simonelli, Manuel Lopez-Antequera
Publikováno v:
ICCV
In this paper we propose an approach for monocular 3D object detection from a single RGB image, which leverages a novel disentangling transformation for 2D and 3D detection losses and a novel, self-supervised confidence score for 3D bounding boxes. O
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::e88ab77090486e560253abfc58befe5e
https://doi.org/10.1109/iccv.2019.00208
https://doi.org/10.1109/iccv.2019.00208