InvisibiliTee: Angle-agnostic Cloaking from Person-Tracking Systems with a Tee

TL;DR

InvisibiliTee introduces a novel angle-agnostic physical adversarial pattern on T-shirts that effectively fools person-tracking systems from all camera angles, enhancing wearer privacy.

Contribution

The paper presents a new black-box adversarial attack method with an angle-agnostic learning scheme for physical cloaking against person detectors.

Findings

Significant reduction in detection accuracy with InvisibiliTee in digital tests.

Effective physical cloaking demonstrated in real-world environments.

Robustness across unseen detection models and camera angles.

Abstract

After a survey for person-tracking system-induced privacy concerns, we propose a black-box adversarial attack method on state-of-the-art human detection models called InvisibiliTee. The method learns printable adversarial patterns for T-shirts that cloak wearers in the physical world in front of person-tracking systems. We design an angle-agnostic learning scheme which utilizes segmentation of the fashion dataset and a geometric warping process so the adversarial patterns generated are effective in fooling person detectors from all camera angles and for unseen black-box detection models. Empirical results in both digital and physical environments show that with the InvisibiliTee on, person-tracking systems' ability to detect the wearer drops significantly.