Measuring Similarity in the Semantic Representation of Moving Objects in Video.
ABSTRACT There are more and more researchers concentrate on the spatio-temporal relationships during the video retrieval process. However,
these researches are just limited to trajectory-based or content-based retrieval, and we seldom retrieve information referring
to semantics. For satisfying the naive users’ requirement from the common point of view, in this paper, we propose a novel
approach for motion recognition from the aspect of semantic meaning. This issue can be addressed through a hierarchical model
that explains how the human language interacts with motions. And, in the experiment part, we evaluate our new approach using
trajectory distance based on spatial relations to distinguish the conceptual similarity and get the satisfactory results.
- SourceAvailable from: psu.edu[show abstract] [hide abstract]
ABSTRACT: WordNet is an on-line lexical reference system whose design is inspired by current psycholinguistic theories of human lexical memory. English nouns, verbs, and adjectives are organized into synonym sets, each representing one underlying lexical concept. Different relations link the synonym sets.01/1991;
Conference Proceeding: Spatio-temporal representation and retrieval using moving object's trajectories.[show abstract] [hide abstract]
ABSTRACT: In this paper, we propose a new spatio-temporal representation scheme using moving objects' trajectories in video data. In order to support content-based retrieval on video data very well, our representation scheme considers the moving distance of an object during a given time interval as well as its temporal and spatial relations. Based on our representation scheme, we present a new similarity measure algorithms for the trajectory of moving objects, which provides ranking for the retrieved video results. Finally, we show from our experiment that our representation scheme achieves about 20% higher precision while holding about the same recall, compared with Li's and Shan's schemes.Proceedings of the ACM Multimedia 2000 Workshops, Los Angeles, CA, USA, October 30 - November 3, 2000; 01/2000
Conference Proceeding: Modeling and querying videos by content trajectories[show abstract] [hide abstract]
ABSTRACT: We present a hierarchical approach to model video shots in three levels: object level (OL), frame level (FL), and shot level (SL). The model captures the visual features of individual objects at OL, visual-spatio-temporal (VST) relationships between objects at FL, and time-varying visual features and time-varying VST relationships at SL. We call the combination of the time-varying visual features and the time-varying VST relationships a content trajectory that is used to represent and index a video shot. A novel query interface, that allows users to describe queries, by sketch and feature specification, is presented. Our experimental results prove the effectiveness of modeling and querying video shots using the content trajectory approachMultimedia and Expo, 2000. ICME 2000. 2000 IEEE International Conference on; 02/2000