MAP: Multispectral Adversarial Patch to Attack Person Detection

Cited 6 time in webofscience Cited 0 time in scopus
  • Hit : 90
  • Download : 0
DC FieldValueLanguage
dc.contributor.authorKim, Taeheonko
dc.contributor.authorLee, Hong Jooko
dc.contributor.authorRo, Yong Manko
dc.date.accessioned2022-11-18T02:01:40Z-
dc.date.available2022-11-18T02:01:40Z-
dc.date.created2022-01-24-
dc.date.created2022-01-24-
dc.date.created2022-01-24-
dc.date.created2022-01-24-
dc.date.issued2022-05-23-
dc.identifier.citation47th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022, pp.4853 - 4857-
dc.identifier.issn1520-6149-
dc.identifier.urihttp://hdl.handle.net/10203/299901-
dc.description.abstractRecently, multispectral person detection has shown great performance in real world applications such as autonomous driving and security systems. However, the reliability of person detection against physical attacks has not been fully explored yet in multispectral person detectors. To evaluate the robustness of multispectral person detectors in the physical world, we propose a novel Multispectral Adversarial Patch (MAP) generation framework. MAP is optimized with a Cross-spectral Mapping(CSM) and Material Emissivity(ME) loss. This paper is the first to evaluate the reliability of a multispectral person detector against physical attack. Throughout experiment, our proposed adversarial patch successfully attacks the person detector and the Average Precision (AP) score is dropped by 90.79% in digital space and 73.34% in physical space.-
dc.languageEnglish-
dc.publisherIEEE Signal Processing Society-
dc.titleMAP: Multispectral Adversarial Patch to Attack Person Detection-
dc.typeConference-
dc.identifier.wosid000864187905029-
dc.identifier.scopusid2-s2.0-85134021630-
dc.type.rimsCONF-
dc.citation.beginningpage4853-
dc.citation.endingpage4857-
dc.citation.publicationname47th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022-
dc.identifier.conferencecountrySI-
dc.identifier.conferencelocationVirtual-
dc.identifier.doi10.1109/ICASSP43922.2022.9747896-
dc.contributor.localauthorRo, Yong Man-
dc.contributor.nonIdAuthorKim, Taeheon-
dc.contributor.nonIdAuthorLee, Hong Joo-
Appears in Collection
EE-Conference Papers(학술회의논문)
Files in This Item
There are no files associated with this item.
This item is cited by other documents in WoS
⊙ Detail Information in WoSⓡ Click to see webofscience_button
⊙ Cited 6 items in WoS Click to see citing articles in records_button

qr_code

  • mendeley

    citeulike


rss_1.0 rss_2.0 atom_1.0