Weakness evaluation on in-vehicle violence detection: an assessment of X3D, C2D and I3D against FGSM and PGD
Autor(a) principal: | |
---|---|
Data de Publicação: | 2022 |
Outros Autores: | , , , , |
Tipo de documento: | Artigo |
Idioma: | eng |
Título da fonte: | Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
Texto Completo: | https://hdl.handle.net/1822/78013 |
Resumo: | When constructing a deep learning model for recognizing violence inside a vehicle, it is crucial to consider several aspects. One aspect is the computational limitations, and the other is the deep learning model architecture chosen. Nevertheless, to choose the best deep learning model, it is necessary to test and evaluate the model against adversarial attacks. This paper presented three different architecture models for violence recognition inside a vehicle. These model architectures were evaluated based on adversarial attacks and interpretability methods. An analysis of the model’s convergence was conducted, followed by adversarial robustness for each model and a sanity-check based on interpretability analysis. It compared a standard evaluation for training and testing data samples with the adversarial attacks techniques. These two levels of analysis are essential to verify model weakness and sensibility regarding the complete video and in a frame-by-frame way. |
id |
RCAP_722ab4e8edf1b4060deeda3a862d9706 |
---|---|
oai_identifier_str |
oai:repositorium.sdum.uminho.pt:1822/78013 |
network_acronym_str |
RCAP |
network_name_str |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
repository_id_str |
7160 |
spelling |
Weakness evaluation on in-vehicle violence detection: an assessment of X3D, C2D and I3D against FGSM and PGDAction recognitionDeep learningIn-car recognitionViolence recognitionScience & TechnologyWhen constructing a deep learning model for recognizing violence inside a vehicle, it is crucial to consider several aspects. One aspect is the computational limitations, and the other is the deep learning model architecture chosen. Nevertheless, to choose the best deep learning model, it is necessary to test and evaluate the model against adversarial attacks. This paper presented three different architecture models for violence recognition inside a vehicle. These model architectures were evaluated based on adversarial attacks and interpretability methods. An analysis of the model’s convergence was conducted, followed by adversarial robustness for each model and a sanity-check based on interpretability analysis. It compared a standard evaluation for training and testing data samples with the adversarial attacks techniques. These two levels of analysis are essential to verify model weakness and sensibility regarding the complete video and in a frame-by-frame way.This work is funded by “FCT—Fundação para a Ciência e Tecnologia” within the R&D Units Project Scope: UIDB/00319/2020. The employment contract of Dalila Durães is supported by CCDR-N Project: NORTE-01-0145-FEDER-000086MDPIUniversidade do MinhoSantos, FlávioDurães, DalilaMarcondes, Francisco S.Hammerschmidt, NiklasMachado, José ManuelNovais, Paulo2022-032022-03-01T00:00:00Zinfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/articleapplication/pdfhttps://hdl.handle.net/1822/78013engSantos, F.; Durães, D.; Marcondes, F.S.; Hammerschmidt, N.; Machado, J.; Novais, P. Weakness Evaluation on In-Vehicle Violence Detection: An Assessment of X3D, C2D and I3D against FGSM and PGD. Electronics 2022, 11, 852. https://doi.org/10.3390/electronics110608522079-929210.3390/electronics11060852https://www.mdpi.com/2079-9292/11/6/852info:eu-repo/semantics/openAccessreponame:Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)instname:Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informaçãoinstacron:RCAAP2023-12-02T01:19:57Zoai:repositorium.sdum.uminho.pt:1822/78013Portal AgregadorONGhttps://www.rcaap.pt/oai/openaireopendoar:71602024-03-19T19:22:34.745615Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) - Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informaçãofalse |
dc.title.none.fl_str_mv |
Weakness evaluation on in-vehicle violence detection: an assessment of X3D, C2D and I3D against FGSM and PGD |
title |
Weakness evaluation on in-vehicle violence detection: an assessment of X3D, C2D and I3D against FGSM and PGD |
spellingShingle |
Weakness evaluation on in-vehicle violence detection: an assessment of X3D, C2D and I3D against FGSM and PGD Santos, Flávio Action recognition Deep learning In-car recognition Violence recognition Science & Technology |
title_short |
Weakness evaluation on in-vehicle violence detection: an assessment of X3D, C2D and I3D against FGSM and PGD |
title_full |
Weakness evaluation on in-vehicle violence detection: an assessment of X3D, C2D and I3D against FGSM and PGD |
title_fullStr |
Weakness evaluation on in-vehicle violence detection: an assessment of X3D, C2D and I3D against FGSM and PGD |
title_full_unstemmed |
Weakness evaluation on in-vehicle violence detection: an assessment of X3D, C2D and I3D against FGSM and PGD |
title_sort |
Weakness evaluation on in-vehicle violence detection: an assessment of X3D, C2D and I3D against FGSM and PGD |
author |
Santos, Flávio |
author_facet |
Santos, Flávio Durães, Dalila Marcondes, Francisco S. Hammerschmidt, Niklas Machado, José Manuel Novais, Paulo |
author_role |
author |
author2 |
Durães, Dalila Marcondes, Francisco S. Hammerschmidt, Niklas Machado, José Manuel Novais, Paulo |
author2_role |
author author author author author |
dc.contributor.none.fl_str_mv |
Universidade do Minho |
dc.contributor.author.fl_str_mv |
Santos, Flávio Durães, Dalila Marcondes, Francisco S. Hammerschmidt, Niklas Machado, José Manuel Novais, Paulo |
dc.subject.por.fl_str_mv |
Action recognition Deep learning In-car recognition Violence recognition Science & Technology |
topic |
Action recognition Deep learning In-car recognition Violence recognition Science & Technology |
description |
When constructing a deep learning model for recognizing violence inside a vehicle, it is crucial to consider several aspects. One aspect is the computational limitations, and the other is the deep learning model architecture chosen. Nevertheless, to choose the best deep learning model, it is necessary to test and evaluate the model against adversarial attacks. This paper presented three different architecture models for violence recognition inside a vehicle. These model architectures were evaluated based on adversarial attacks and interpretability methods. An analysis of the model’s convergence was conducted, followed by adversarial robustness for each model and a sanity-check based on interpretability analysis. It compared a standard evaluation for training and testing data samples with the adversarial attacks techniques. These two levels of analysis are essential to verify model weakness and sensibility regarding the complete video and in a frame-by-frame way. |
publishDate |
2022 |
dc.date.none.fl_str_mv |
2022-03 2022-03-01T00:00:00Z |
dc.type.status.fl_str_mv |
info:eu-repo/semantics/publishedVersion |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/article |
format |
article |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
https://hdl.handle.net/1822/78013 |
url |
https://hdl.handle.net/1822/78013 |
dc.language.iso.fl_str_mv |
eng |
language |
eng |
dc.relation.none.fl_str_mv |
Santos, F.; Durães, D.; Marcondes, F.S.; Hammerschmidt, N.; Machado, J.; Novais, P. Weakness Evaluation on In-Vehicle Violence Detection: An Assessment of X3D, C2D and I3D against FGSM and PGD. Electronics 2022, 11, 852. https://doi.org/10.3390/electronics11060852 2079-9292 10.3390/electronics11060852 https://www.mdpi.com/2079-9292/11/6/852 |
dc.rights.driver.fl_str_mv |
info:eu-repo/semantics/openAccess |
eu_rights_str_mv |
openAccess |
dc.format.none.fl_str_mv |
application/pdf |
dc.publisher.none.fl_str_mv |
MDPI |
publisher.none.fl_str_mv |
MDPI |
dc.source.none.fl_str_mv |
reponame:Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) instname:Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação instacron:RCAAP |
instname_str |
Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação |
instacron_str |
RCAAP |
institution |
RCAAP |
reponame_str |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
collection |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
repository.name.fl_str_mv |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) - Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação |
repository.mail.fl_str_mv |
|
_version_ |
1799132697137774592 |