
Sign up to save your podcasts
Or
Wiebe van Ranst joins us to talk about a project in which specially designed printed images can fool a computer vision system, preventing it from identifying a person. Their attack targets the popular YOLO2 pre-trained image recognition model, and thus, is likely to be widely applicable.
4.4
473473 ratings
Wiebe van Ranst joins us to talk about a project in which specially designed printed images can fool a computer vision system, preventing it from identifying a person. Their attack targets the popular YOLO2 pre-trained image recognition model, and thus, is likely to be widely applicable.
585 Listeners
624 Listeners
298 Listeners
340 Listeners
140 Listeners
770 Listeners
270 Listeners
183 Listeners
63 Listeners
298 Listeners
91 Listeners
105 Listeners
201 Listeners
72 Listeners
496 Listeners