Zobrazeno 1 - 10
of 41
pro vyhledávání: '"Patrick Cardinal"'
Autor:
Mohammad Esmaeilpour, Nourhene Chaalia, Adel Abusitta, Franşois-Xavier Devailly, Wissem Maazoun, Patrick Cardinal
Publikováno v:
Pattern Recognition Letters. 159:204-210
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::cc5675d5d0a34b281f4d54e6a80302d1
https://doi.org/10.1016/j.patrec.2022.05.023
https://doi.org/10.1016/j.patrec.2022.05.023
Publikováno v:
IEEE Transactions on Information Forensics and Security. 17:2044-2058
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::0ebc6a7c1ad84b92c7396e8b9b24d396
https://doi.org/10.1109/tifs.2022.3175603
https://doi.org/10.1109/tifs.2022.3175603
Publikováno v:
IEEE Transactions on Information Forensics and Security. 15:2147-2159
Adversarial audio attacks can be considered as a small perturbation unperceptive to human ears that is intentionally added to the audio signal and causes a machine learning model to make mistakes. This poses a security concern about the safety of mac
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::bc7bc409a8da7dda02808563f85dec5d
https://doi.org/10.1109/tifs.2019.2956591
https://doi.org/10.1109/tifs.2019.2956591
This paper introduces a new synthesis-based defense algorithm for counteracting with a varieties of adversarial attacks developed for challenging the performance of the cutting-edge speech-to-text transcription systems. Our algorithm implements a Sob
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::469ac8712a52cadadca5b1bdb0ffcbd4
This paper proposes a new defense approach for counteracting state-of-the-art white and black-box adversarial attack algorithms. Our approach fits into the implicit reactive defense algorithm category since it does not directly manipulate the potenti
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::0c0cf6f3c350a0c9582ac18fb7f7d342
http://arxiv.org/abs/2103.14717
http://arxiv.org/abs/2103.14717
Multimodal analysis has recently drawn much interest in affective computing, since it can improve the overall accuracy of emotion recognition over isolated uni-modal approaches. The most effective techniques for multimodal emotion recognition efficie
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::83d53096ad095864903969047109a67c
This paper introduces a novel adversarial algorithm for attacking the state-of-the-art speech-to-text systems, namely DeepSpeech, Kaldi, and Lingvo. Our approach is based on developing an extension for the conventional distortion condition of the adv
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::6b87ec6ab8e82d313a56c042115c45c8
RADARSAT-2 Synthetic-Aperture Radar Land Cover Segmentation Using Deep Convolutional Neural Networks
Publikováno v:
Pattern Recognition. ICPR International Workshops and Challenges ISBN: 9783030687922
ICPR Workshops (8)
ICPR Workshops (8)
Synthetic Aperture Radar (SAR) imagery captures the physical properties of the Earth by transmitting microwave signals to its surface and analyzing the backscattered signal. It does not depends on sunlight and therefore can be obtained in any conditi
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::4cfd6cca67427f2d9ed0b0726baf719d
https://doi.org/10.1007/978-3-030-68793-9_8
https://doi.org/10.1007/978-3-030-68793-9_8
Publikováno v:
FG
Automatic pain assessment has an important potential diagnostic value for populations that are incapable of articulating their pain experiences. As one of the dominating nonverbal channels for eliciting pain expression events, facial expressions has
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::16bab817b09ecbd5620d0254c822b8ed
https://doi.org/10.1109/fg47880.2020.00139
https://doi.org/10.1109/fg47880.2020.00139
Publikováno v:
ICASSP
In this paper we propose a novel defense approach against end-to-end adversarial attacks developed to fool advanced speech-to-text systems such as DeepSpeech and Lingvo. Unlike conventional defense approaches, the proposed approach does not directly
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::6e2baed1fe37b5cfac741f56ee4d3e28
http://arxiv.org/abs/2010.11352
http://arxiv.org/abs/2010.11352