按年被提到
20192019 · 1 次2026
AI 安全、伦理与社会影响›算法偏见与监控·arXiv:1910.14667 · cs.CV

Making an Invisibility Cloak: Real World Adversarial Attacks on Object Detectors

Zuxuan Wu, Ser-Nam Lim, Larry Davis, Tom Goldstein · 2019
开始阅读 ↗PDFalphaXiv 讨论
被提及 · 2 人首次 2019-11-16 · 最近 2022-10-28
2不同的人
2活跃月份
2讨论串
243被引(OpenAlex)
arXiv 摘要(候选层不写导读)

We present a systematic study of adversarial attacks on state-of-the-art object detection frameworks. Using standard detection datasets, we train patterns that suppress the objectness scores produced by a range of commonly used detectors, and ensembles of detectors. Through extensive experiments, we benchmark the effectiveness of adversarially trained patches under both white-box and black-box settings, and quantify transferability of attacks between datasets, object classes, and detector models. Finally, we present a detailed study of physical world attacks using printed posters and wearable clothes, and rigorously quantify the performance of such attacks with different metrics.

候选层:至少 2 个不同的人、在 2 个不同月份提到过,还没到主书架的 3 人 3 个月。没有做立场标注,所以只写“被提及”。

HN 书评

没有评论直接链到它;下面是它被提交的讨论串里回复最多的评论(不计入统计)
讨论串评论rapjr9 · 2022-10

“So does work like this essentially mean there will always be strange edge cases in AI algorithms so that they will never be 100% accurate? Which means the AI driven cars and robots will also never be perfect at navigating the world? What degree of accuracy would we ever be comfortable with? The current metrics for AI seem too simplistic, there needs to be…”

Real World Adversarial Attacks on Object Detectors原评论 ↗
被提交为 story · 2 次