時間：2023/03/03 13:20~13:50
地點：電1F02
演講者：吳志強 博士
Abstact:
Anomaly Detection (AD) aims to discover irregular patterns in spatial or temporal domains while the difference between abnormal and normal fundamentally defines the anomaly. The task is promising and extensively applicable in the field of computer vision. Consequently, developing an algorithm for automatically verifying defects or anomalous events with reliable accuracy is crucial. In this talk, we will share our recent works that comprise three parts: simulated one-class AD, real-world image AD, and video AD, where the developed approaches are independent but focus on various AD.
