Detecting abnormal behaviors in video

Challenges, Machine Learning Solutions, Future Prospective

Anomaly detection is one of the very interesting applications, which relies on machine learning algorithms and domain knowledge. Anomalies refer to a set of data points that are considerably different from the remainder of the data. Anomaly detection, therefore, aims to detect abnormal or suspicious patterns in the data.

In this talk, Huamin Ren from Inmeta Consulting AS will illustrate a specific application of detecting abnormal behaviors in video, which can be regarded as one of anomaly detection applications. She will introduce the challenge of this task, state-of-the-art methods used, and then give insights into future bigger pictures of applying these methods into other anomaly detection applications.

https://secure.meetupstatic.com/photos/event/3/7/7/600_464940887.jpeg