Understanding Complex Activities in Streaming Videos
Speaker: Ehsan Elhamifar, Associate Professor of Computer Sciences, Northeastern University (Boston, USA)
Data Science is relevant for all disciplines, but there is great variation in the application of methods and techniques. Get a taste for what data science is, and let yourself be inspired by how data science methods are put into practice in our Data Science Seminar!
Event details from: Understanding Complex Activities in Streaming Videos
- Date
- 17 July 2024
- Time
- 11:00 – 12:00
- Room
- Doelenzaal
About the seminar presentation
Understanding Complex Activities in Streaming Videos
Humans perform a wide range of complex procedural activities, such as cooking hour-long recipes, assembling and repairing devices and performing medical procedures. Learning such activities from videos allows us to design intelligent task assistants, robots and coaching platforms that perform or guide people through tasks.
In this talk, Ehsan will discuss an intelligent task assistant system that he has been developing in his lab, and present new neural architectures as well as efficient learning and inference frameworks to understand complex activity videos, addressing the following challenges:
- Task videos are long with long-range action dependencies, contain task-irrelevant activities, demonstrate different ways of performing the same task
- At inference time, as data arrive in real-time, we must accurately recognize actions and their progress especially with only a few frames.
- We need to detect possible errors during task executions. However, the space of errors is extremely large and we cannot hope to gather training videos for all of them.
Registration
Everyone from all disciplines is welcome to attend! The presentation will take place in-person only, and there will be an opportunity to ask questions and engage in lively discussion on the day.

About the speaker
Ehsan Elhamifar is an Associate Professor in the Khoury College of Computer Sciences, the director of the Mathematical Data Science (MCADS) Lab and the Director of MS in AI at Northeastern University (Boston, USA).
He has broad research interests in computer vision and machine learning. The overarching goal of his research is to develop computer vision systems that learns from and makes inferences about data analogous to humans. He is a recipient of the DARPA Young Faculty Award. Prior to Northeastern, he was a postdoctoral scholar in the EECS department at UC Berkeley. He obtained his PhD in ECE at the Johns Hopkins University (JHU) and received two Masters degrees, one in EE from Sharif University of Technology in Iran and another in Applied Mathematics and Statistics from JHU.