Detecting Road Obstacles by Erasing Them
IEEE Transactions on Pattern Analysis and Machine Intelligence · 2024-04 · 46 (4) · 2450-2460
DOI: 10.1109/tpami.2023.3335152 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Pattern Analysis and Machine Intelligence, 2024-04. Verfasst von Krzysztof Lis und 3 weiteren Autor:innen. Die Arbeit wurde bisher 34-mal zitiert (Crossref) und nennt 59 Literaturangaben. Gefördert von: International Chair Drive for All - MINES ParisTech - Peugeot-Citroën - Safran - Valeo.
Abstract (Originalsprache)
Vehicles can encounter a myriad of obstacles on the road, and it is impossible to record them all beforehand to train a detector. Instead, we select image patches and inpaint them with the surrounding road texture, which tends to remove obstacles from those patches. We then use a network trained to recognize discrepancies between the original patch and the inpainted one, which signals an erased obstacle.
Förderung
- International Chair Drive for All - MINES ParisTech - Peugeot-Citroën - Safran - Valeo
Zitieren
APA
Lis, K., Honari, S., Fua, P., & Salzmann, M. (2024). Detecting Road Obstacles by Erasing Them. IEEE Transactions on Pattern Analysis and Machine Intelligence, 46(4), 2450-2460. https://doi.org/10.1109/tpami.2023.3335152
Vancouver
Lis K, Honari S, Fua P, Salzmann M. Detecting Road Obstacles by Erasing Them. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2024;46(4):2450-2460. doi:10.1109/tpami.2023.3335152
BibTeX
@article{Lis2024,
author = {Lis, Krzysztof and Honari, Sina and Fua, Pascal and Salzmann, Mathieu},
title = {Detecting Road Obstacles by Erasing Them},
journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence},
year = {2024},
volume = {46},
number = {4},
pages = {2450--2460},
doi = {10.1109/tpami.2023.3335152}
}
EUVIMED