Imaging Science Seminar: Video Event Monitoring for Smart Home
Video Event Monitoring for Smart Home
Hongcheng Wang
Director of Research
Comcast AI
Register Here for Zoom Link
Abstract:
Video Event Monitoring for Smart Home Comcast’s Xfinity Home connects millions of home smart cameras and IoT devices to improve our customers’ safety and security. Our teams use computer vision and deep learning to understand video and sensor data from these devices to identify relevant events so that we can improve the user experience. We developed deep learning based approaches for motion detection, object detection, and activity recognition. Specifically, we developed a relevant motion event detection network (RemoteNet) to detect motion of interest. We also explored the spatial-temporal relationships among objects, scene layout and actions to recognize Agent-in-Place actions. Finally we developed a semi-supervised learning approach for video classification (VideoSSL) to detect certain activities with limited training data. Using these techniques, we have achieved very promising results on relevant event detection and activity recognition for notifications to our customers.
Speaker Bio:
Hongcheng Wang, Director of Research at Comcast AI, leading the xFinity Home Intelligence team and Data Science team. He and his team focus on developing scalable machine/deep learning technologies for video and IoT intelligence and connectivity, and predictive and diagnostic intelligence for customer service. He has published 40+ technical papers, and holds 50+ patents and patent applications (28 issued). He has been serving as an Associate Editor (AE) for the Visual Computer Journal by Springer since 2012, and Area Chairs or industry Chairs for a number of conferences. He holds a PhD in computer engineering from the University of Illinois at Urbana-Champaign (UIUC).
Intended Audience:
Beginners, undergraduates, graduates. Those with interest in the topic.
Event Snapshot
When and Where
Who
This is an RIT Only Event
Interpreter Requested?
No