Biblio
Filters: Author is Seungwon Lee [Clear All Filters]
.
2014. Object retrieval using scene normalized human model for video surveillance system. Consumer Electronics (ISCE 2014), The 18th IEEE International Symposium on. :1-2.
This paper presents a human model-based feature extraction method for a video surveillance retrieval system. The proposed method extracts, from a normalized scene, object features such as height, speed, and representative color using a simple human model based on multiple-ellipse. Experimental results show that the proposed system can effectively track moving routes of people such as a missing child, an absconder, and a suspect after events.