-
Notifications
You must be signed in to change notification settings - Fork 23
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
- Loading branch information
Showing
2 changed files
with
14 additions
and
1 deletion.
There are no files selected for viewing
13 changes: 13 additions & 0 deletions
13
2024/2024_10_28_Adversarial_attacks_against_object_detection.md
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,13 @@ | ||
# Adversarial examples vs. context consistency defense for object detection | ||
|
||
Slides: https://hbaniecki.com/mi2seminar2024 | ||
|
||
## Abstract | ||
|
||
One defense strategy for detecting adversarial examples is to check for intrinsic context consistencies in the input data, where context refers to various relationships (e.g., object-to-object co-occurrence relationships) in images. I will present a paper showing that these context consistency checks don't work for properly crafted adversarial examples. ADC defines a joint optimization problem with two attack goals: (1) fooling the object detector and (2) evading the context consistency check system, at the same time. Experiments on PASCAL VOC and MS COCO datasets show that examples generated with ADC fool the object detector with a success rate of over 85%, and at the same time evade the recently proposed context consistency checks, with a “bypassing” rate of over 80%. | ||
|
||
## Source papers | ||
|
||
[Connecting the Dots: Detecting Adversarial Perturbations Using Context Inconsistency](https://arxiv.org/abs/2007.09763) | ||
|
||
[ADC: Adversarial attacks against object Detection that evade Context consistency checks](https://arxiv.org/abs/2110.12321) |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters