Adversarial Patch · HackerPapers discussions

Adversarial Patch — HN submission

Fri, 13 Feb 2026 18:16:20 GMT

Adversarial Patch: images that make classifiers ignore other items in a scene

1 point; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.

Adversarial Patch — mentioned by lsb

Fri, 22 Mar 2024 22:15:55 GMT

Yes, these are “adversarial patches” in image classification, like https://arxiv.org/abs/1712.09665 . Similarly you can take these adversaries and add them to your own larger model, in an arms race.

Paper and discussion history. A mention is not an endorsement.

Adversarial Patch — mentioned by BelleOfTheBall

Fri, 10 Apr 2020 21:47:39 GMT

Honestly, I'd wear that dress easily, even for regular outings, not just for hiding from cameras. Those hoodies, though, maybe I'm an old fart but they're a bit too "abstract". Sadly, those seem like the most effective option right now. I read the Adversarial Patch article [0] and it pretty much demonstrated similar patterns. Trouble is, if these get widespread enough - they won't look out of place but the moment they become widespread is the moment updates get pushed to make them obsolete. It's an unescapable situation. [0]: https://arxiv.org/abs/1712.09665

Paper and discussion history. A mention is not an endorsement.

Adversarial Patch — mentioned by lwneal

Sun, 12 Jan 2020 21:20:31 GMT

The article mentions the use of "Juggalo" style face paint to avoid face detection. This type of countermeasure is effective against conventional HOG-SVM type face detectors such as [1], but would be inadvisable for daily use. To deal with more advanced convolutional detectors, consider wearing a hat or clothing with adversarial patches [2], simple patterns that can be printed onto stickers. These patches work by feeding the detector inputs that cause an overwhelming bias toward a particular class (eg. toaster), which drowns out the real signal (eg. human face). Even the first generation of…

Paper and discussion history. A mention is not an endorsement.

Adversarial Patch — mentioned by mrkoot

Fri, 23 Aug 2019 19:48:48 GMT

For better or worse, this helps establish neural networks that are not (or less?) vulnerable to deception via e.g. fake eyeglasses [ LINK: https://dl.acm.org/citation.cfm?id=2978392 ] and adversarial stickers [ LINK: https://arxiv.org/abs/1712.09665 ]. From the post: "We’ve developed a method to assess whether a neural network classifier can reliably defend against adversarial attacks not seen during training [such as Elastic, Fog, Gabor, and Snow]. Our method yields a new metric, UAR (Unforeseen Attack Robustness), which evaluates the robustness of a single model against an unanticipated…

Paper and discussion history. A mention is not an endorsement.

Adversarial Patch — HN submission

Wed, 05 Dec 2018 06:20:54 GMT

Ask HN: Has anyone used adversarial patches for effect?

Brown’s adversarial patch paper came out in May [0] and was discussed a bit on HN [1]. But I’m wondering if anyone has used or seen reporting on use of these patches. They seem counterproductive digitally as, while difficult by algorithm, should be easy to flag for human review. So like wearing a sticker saying “I don’t want to be understood.” But in real life seem useful for avoiding automatic detection as part of a crowd. I’m waiting to see these pop up like in events like the recent Paris tax riots. [0] https://arxiv.org/pdf/1712.09665.pdf https://arxiv.org/abs/1712.09665 [1]…

2 points; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.

Adversarial Patch — mentioned by dpflan

Thu, 24 May 2018 16:03:35 GMT

Masks, face altering, stickers, etc. I like this research about "Adversarial Patches" - stickers that confuse image recognition systems. I imagine one may wear a display with ever changing images that are salient and confuse such systems. > Research Article: https://arxiv.org/pdf/1712.09665.pdf > Video for patches: https://www.youtube.com/watch?v=i1sp4X57TL4

Paper and discussion history. A mention is not an endorsement.

Adversarial Patch — HN submission

Tue, 09 Jan 2018 11:51:58 GMT

Adversarial Patch [pdf]

34 points; 3 discussion replies on HN. Paper and source history. A submission is not an endorsement.

Adversarial Patch — HN submission

Wed, 03 Jan 2018 15:11:23 GMT

Google Research team: adversial patches to confuse Deep Learning algorithms

1 point; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.

Adversarial Patch — HN submission

Wed, 03 Jan 2018 14:15:42 GMT

Adversarial Patch

7 points; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.

Adversarial Patch — HN submission

Sat, 30 Dec 2017 22:58:33 GMT

Adversarial Patch [pdf]

2 points; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.