Anomaly Detection at 30% anomaly

In the field of computer vision, anomaly detection is a key technology aimed at identifying abnormal patterns in data. This task focuses on unsupervised anomaly detection with a 30% anomaly rate, which aims to accurately identify the 30% of samples that are anomalies within an unlabelled dataset. The application value of this task lies in its ability to effectively enhance the system's response to abnormal events, making it suitable for scenarios such as industrial monitoring, medical diagnosis, and cybersecurity.