Matches in SemOpenAlex for { <https://semopenalex.org/work/W3204321633> ?p ?o ?g. }
Showing items 1 to 100 of
100
with 100 items per page.
- W3204321633 abstract "Blind spots or outright deceit can bedevil and deceive machine learning models. Unidentified objects such as digital stickers, also known as adversarial patches, can fool facial recognition systems, surveillance systems and self-driving cars. Fortunately, most existing adversarial patches can be outwitted, disabled and rejected by a simple classification network called an adversarial patch detector, which distinguishes adversarial patches from original images. An object detector classifies and predicts the types of objects within an image, such as by distinguishing a motorcyclist from the motorcycle, while also localizing each object's placement within the image by drawing so-called bounding boxes around each object, once again separating the motorcyclist from the motorcycle. To train detectors even better, however, we need to keep subjecting them to confusing or deceitful adversarial patches as we probe for the models' blind spots. For such probes, we came up with a novel approach, a Low-Detectable Adversarial Patch, which attacks an object detector with small and texture-consistent adversarial patches, making these adversaries less likely to be recognized. Concretely, we use several geometric primitives to model the shapes and positions of the patches. To enhance our attack performance, we also assign different weights to the bounding boxes in terms of loss function. Our experiments on the common detection dataset COCO as well as the driving-video dataset D2-City show that LDAP is an effective attack method, and can resist the adversarial patch detector." @default.
- W3204321633 created "2021-10-11" @default.
- W3204321633 creator A5015663174 @default.
- W3204321633 creator A5024608436 @default.
- W3204321633 creator A5025588570 @default.
- W3204321633 creator A5063906798 @default.
- W3204321633 creator A5086215557 @default.
- W3204321633 date "2021-09-30" @default.
- W3204321633 modified "2023-09-23" @default.
- W3204321633 title "You Cannot Easily Catch Me: A Low-Detectable Adversarial Patch for Object Detectors" @default.
- W3204321633 cites W1536680647 @default.
- W3204321633 cites W1686810756 @default.
- W3204321633 cites W1861492603 @default.
- W3204321633 cites W2102605133 @default.
- W3204321633 cites W2134104018 @default.
- W3204321633 cites W2174349387 @default.
- W3204321633 cites W2194775991 @default.
- W3204321633 cites W2243397390 @default.
- W3204321633 cites W2565639579 @default.
- W3204321633 cites W2570343428 @default.
- W3204321633 cites W2604505099 @default.
- W3204321633 cites W2613718673 @default.
- W3204321633 cites W2796347433 @default.
- W3204321633 cites W2798302089 @default.
- W3204321633 cites W2950141105 @default.
- W3204321633 cites W2962700793 @default.
- W3204321633 cites W2962748759 @default.
- W3204321633 cites W2962785568 @default.
- W3204321633 cites W2963037989 @default.
- W3204321633 cites W2963150697 @default.
- W3204321633 cites W2963207607 @default.
- W3204321633 cites W2963542991 @default.
- W3204321633 cites W2963557656 @default.
- W3204321633 cites W2963726920 @default.
- W3204321633 cites W2964175514 @default.
- W3204321633 cites W2964253222 @default.
- W3204321633 cites W2965198951 @default.
- W3204321633 cites W3017485054 @default.
- W3204321633 cites W3018757597 @default.
- W3204321633 cites W3033943140 @default.
- W3204321633 cites W3094984771 @default.
- W3204321633 cites W3097573595 @default.
- W3204321633 cites W3106250896 @default.
- W3204321633 hasPublicationYear "2021" @default.
- W3204321633 type Work @default.
- W3204321633 sameAs 3204321633 @default.
- W3204321633 citedByCount "0" @default.
- W3204321633 crossrefType "posted-content" @default.
- W3204321633 hasAuthorship W3204321633A5015663174 @default.
- W3204321633 hasAuthorship W3204321633A5024608436 @default.
- W3204321633 hasAuthorship W3204321633A5025588570 @default.
- W3204321633 hasAuthorship W3204321633A5063906798 @default.
- W3204321633 hasAuthorship W3204321633A5086215557 @default.
- W3204321633 hasConcept C153180895 @default.
- W3204321633 hasConcept C154945302 @default.
- W3204321633 hasConcept C2776151529 @default.
- W3204321633 hasConcept C2781238097 @default.
- W3204321633 hasConcept C31972630 @default.
- W3204321633 hasConcept C37736160 @default.
- W3204321633 hasConcept C41008148 @default.
- W3204321633 hasConcept C63584917 @default.
- W3204321633 hasConcept C76155785 @default.
- W3204321633 hasConcept C94915269 @default.
- W3204321633 hasConceptScore W3204321633C153180895 @default.
- W3204321633 hasConceptScore W3204321633C154945302 @default.
- W3204321633 hasConceptScore W3204321633C2776151529 @default.
- W3204321633 hasConceptScore W3204321633C2781238097 @default.
- W3204321633 hasConceptScore W3204321633C31972630 @default.
- W3204321633 hasConceptScore W3204321633C37736160 @default.
- W3204321633 hasConceptScore W3204321633C41008148 @default.
- W3204321633 hasConceptScore W3204321633C63584917 @default.
- W3204321633 hasConceptScore W3204321633C76155785 @default.
- W3204321633 hasConceptScore W3204321633C94915269 @default.
- W3204321633 hasLocation W32043216331 @default.
- W3204321633 hasOpenAccess W3204321633 @default.
- W3204321633 hasPrimaryLocation W32043216331 @default.
- W3204321633 hasRelatedWork W2296054617 @default.
- W3204321633 hasRelatedWork W2608797168 @default.
- W3204321633 hasRelatedWork W2811414481 @default.
- W3204321633 hasRelatedWork W2885106262 @default.
- W3204321633 hasRelatedWork W2917575296 @default.
- W3204321633 hasRelatedWork W2943520232 @default.
- W3204321633 hasRelatedWork W2946952476 @default.
- W3204321633 hasRelatedWork W2962846186 @default.
- W3204321633 hasRelatedWork W3003990034 @default.
- W3204321633 hasRelatedWork W3014741892 @default.
- W3204321633 hasRelatedWork W3037534284 @default.
- W3204321633 hasRelatedWork W3043819440 @default.
- W3204321633 hasRelatedWork W3108281670 @default.
- W3204321633 hasRelatedWork W3138480145 @default.
- W3204321633 hasRelatedWork W3192004794 @default.
- W3204321633 hasRelatedWork W3206395574 @default.
- W3204321633 hasRelatedWork W2280068342 @default.
- W3204321633 hasRelatedWork W2975490202 @default.
- W3204321633 hasRelatedWork W3020404794 @default.
- W3204321633 hasRelatedWork W3110243870 @default.
- W3204321633 isParatext "false" @default.
- W3204321633 isRetracted "false" @default.
- W3204321633 magId "3204321633" @default.
- W3204321633 workType "article" @default.