FRAS: A Symbolization-based Approach to Video Similarity Search
Emily Zhou (CSIRO)
CSIRO ICTDATE: 2008-06-24
TIME: 16:00:00 - 17:00:00
LOCATION: CSIT Seminar Room, N101
CONTACT: JavaScript must be enabled to display this email address.
ABSTRACT:
Efficiently and effectively searching for similar videos is an important and non-trivial problem in content-based video search systems. In this talk, we present a symbolization- based approach, namely FRAS, for content-based search on very large video databases. The novelty of FRAS is that it explores the data distribution in subspaces to build a visual dictionary with which the video data are processed by deriving the string matching techniques with two-step data simplification. Specifically, we first propose an adaptive approach called as VLP to divide the whole feature space into a series of subspaces of variable lengthes, from which the dominant ones are selected. By clustering the video keyframes over each dominant subspace, a stable visual dictionary is built and a compact video representation model is developed by transforming each keyframe into a word that is a series of symbols, and further each video into a sequence of words. Then, we present an innovative similarity measure called VED, which adopts a complementary information compensation scheme based on the visual features and sequence context of videos. Finally, an efficient two- layered index strategy based on sequence decomposition and reconstruction and a number of query optimization techniques are proposed to facilitate video similarity search. We demonstrate the high effectiveness and efficiency of FRAS by comparing with the existing video search approaches.
BIO:
Xiangmin Emily Zhou received the B.Sc. Degree in Computer Science from the Faculty of Computers, Shenyang Architecture and Civil Engineering University, China, in 1998. From 1998 to 2000, she worked as a sofeware engineer in Shenzhen Weihao Multimedia Network Group Corporation, China and Shenyang DongYu Group Corporation, China respectively. In Sept. 2000, she started her master study in Institute of Computer System, School of Information and Engineering, Northeastern Unversity, China, and received her master degree in March, 2003. During her master, she won the best student paper award at Australian Database Conference 2003. In 2004, she obtained the IRPS/UQLAS scholarships for her PhD and started her study at University of Queensland from Feb 2005. Currently, she has finished her PhD and joined CSIRO as a Postdoctoral fellow.
