Fooling Computer Vision
Data Skeptic
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.Next Episodes
Algorithmic Fairness @ Data Skeptic
📆 2020-01-14 01:00
Interpretability @ Data Skeptic
📆 2020-01-07 01:00
NLP in 2019 @ Data Skeptic
📆 2019-12-31 01:00
The Limits of NLP @ Data Skeptic
📆 2019-12-24 01:00
Jumpstart Your ML Project @ Data Skeptic
📆 2019-12-15 01:00