Zobrazeno 1 - 10
of 13
pro vyhledávání: '"Marin Oršić"'
Publikováno v:
Remote Sensing, Vol 15, Iss 8, p 1968 (2023)
Dense panoptic prediction is a key ingredient in many existing applications such as autonomous driving, automated warehouses, or remote sensing. Many of these applications require fast inference over large input resolutions on affordable or even embe
Externí odkaz:
https://doaj.org/article/a3f736253516449d81ce02241ff5edfc
Publikováno v:
Sensors, Vol 23, Iss 2, p 940 (2023)
Semi-supervised learning is an attractive technique in practical deployments of deep models since it relaxes the dependence on labeled data. It is especially important in the scope of dense prediction because pixel-level annotation requires substanti
Externí odkaz:
https://doaj.org/article/4333a27bfbfc48f698d3bb6982fff0a4
Publikováno v:
MVA
Semi-supervised learning is especially interesting in the dense prediction context due to high cost of pixel-level ground truth. Unfortunately, most such approaches are evaluated on outdated architectures which hamper research due to very slow traini
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::78f0ef7a9c13911838f51defd620cc26
https://doi.org/10.23919/mva51890.2021.9511402
https://doi.org/10.23919/mva51890.2021.9511402
Autor:
Marin Oršić
Emergence of large datasets and resilience of convolutional models have enabled successful training of very large semantic segmentation models. However, high capacity implies high computational complexity and therefore hinders real-time operation. We
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=57a035e5b1ae::16c418970659cb6819b1ba0c206224b5
https://www.bib.irb.hr/1171013
https://www.bib.irb.hr/1171013
Publikováno v:
ITSC
We address the automatic recognition of road safety attributes according to the iRAP methodology. We formulate the problem as a separate multi-class classification of each iRAP attribute in georeferenced video clips that correspond to particular road
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::c4d83859c34e793cfd7dfeb0cf58b53b
https://doi.org/10.1109/itsc45102.2020.9294305
https://doi.org/10.1109/itsc45102.2020.9294305
Publikováno v:
CVPR
We address anticipation of scene development by forecasting semantic segmentation of future frames. Several previous works approach this problem by F2F (feature-to-feature) forecasting where future features are regressed from observed features. Diffe
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::c6b644e97864c71d27294d5477755396
https://doi.org/10.1109/cvpr42600.2020.01066
https://doi.org/10.1109/cvpr42600.2020.01066
Publikováno v:
Lecture Notes in Computer Science ISBN: 9783030336752
GCPR
GCPR
Future anticipation is of vital importance in autonomous driving and other decision-making systems. We present a method to anticipate semantic segmentation of future frames in driving scenarios based on feature-to-feature forecasting. Our method is b
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::352f07ce162bed55fb9ea07875904c3b
http://arxiv.org/abs/1907.11475
http://arxiv.org/abs/1907.11475
Publikováno v:
Lecture Notes in Computer Science ISBN: 9783030336752
GCPR
GCPR
Recent success on realistic road driving datasets has increased interest in exploring robust performance in real-world applications. One of the major unsolved problems is to identify image content which can not be reliably recognized with a given inf
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::576bb2b2b275c771096f436bcc61a47c
Publikováno v:
CVPR
Recent success of semantic segmentation approaches on demanding road driving datasets has spurred interest in many related application fields. Many of these applications involve real-time prediction on mobile platforms such as cars, drones and variou
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::833a1597ad392fa3d432bdf515912c67
Autor:
Marin Oršić, Siniša Šegvić
Publikováno v:
Pattern Recognition. 110:107611
Emergence of large datasets and resilience of convolutional models have enabled successful training of very large semantic segmentation models. However, high capacity implies high computational complexity and therefore hinders real-time operation. We
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::6e852608a74b4eacf25b5f4d9899b62f
https://doi.org/10.1016/j.patcog.2020.107611
https://doi.org/10.1016/j.patcog.2020.107611