Seminar: Scaled Eigen Appearance Feature and Likelihood Pruning in Large Scale Video Duplicates Search
Presenter: Dr Zhu Li, Senior Research,m Core Networks Research, Huawei Technology, USA
Abstract: The rapid growth of on-line video repositories present new challenges for content management applications, especially detection of robust video duplicates in very large repositories. This talk presents a duplicate likelihood modeling and pruning algorithm, which offers fine flexibility in trade-offs of detection robustness, localisation accuracy and computational complexity. The solution is also efficient in feature compactness. The proposed SEAF video signature has an average data rate of 2-8bits/second, and it can detect and localise video clips of 2-4 minutes within a 116 hour repository within 0.003 seconds on a typical desktop computer.
Brief Bio: Dr Zhu Li (SM '07) is currently a Senior Staff Researcher with Core Networks Research, Huawei Technology, USA at Bridgewater, New Jersey where he leads the Media Analytics Group. He received his PhD in Electrical & Computer Engineering from Northwestern University, Evanston in 2004. He was an Assistant Professor with the Dept of Computing, The Hong Kong Polytechnic University from 2008 to 2010, and a Principal Staff Research Engineer with the Multimedia Research Lab (MRL), Motorola Labs, from 2000 to 2008.
- 17 November 2011
- 14:00 - 15:00
- City - Broadway CB10 Level 4, Room 460
- All Welcome
- Barbara Munday