Abstract
In this paper, we describe the Toshiba event detection system for TRECVID surveillance event detection task [1]. Our system consists of four components: (1) a flexible and robust change detection based on non-parametric background modeling, (2) a human detection that extends and outperforms HOG (Histogram of Oriented Gradient) human detection, (3) a human tracking with simple linear estimation and color histogram matching, and (4) an event detection for three TRECVID required events (E05, E19, and E20) based on change detection and human tracking. Our current system has just adopted a simple version of each component and requires further refinement.
| Original language | English |
|---|---|
| Publication status | Published - 2008 |
| Externally published | Yes |
| Event | TREC Video Retrieval Evaluation, TRECVID 2008 - Gaithersburg, MD, United States Duration: 2008 Nov 17 → 2008 Nov 18 |
Conference
| Conference | TREC Video Retrieval Evaluation, TRECVID 2008 |
|---|---|
| Country/Territory | United States |
| City | Gaithersburg, MD |
| Period | 08/11/17 → 08/11/18 |
ASJC Scopus subject areas
- Computer Graphics and Computer-Aided Design
- Computer Vision and Pattern Recognition
- Human-Computer Interaction
- Software