A NOVEL OBJECT DETECTION TECHNIQUE IN COMPRESSED DOMAIN
檔案
本條目沒有附加檔案。
日期
2006-01-01
類型
系所
期刊名稱
COMPUTER VISION AND GRAPHICS (ICCVG 2004)
卷期與頁次
第 32 卷,頁 689–694
ISBN
978-1-4020-4179-2; 978-1-4020-4178-5
語言
英文
摘要
In this paper we propose a novel approach for robust motion vector based object detection in MPEG-1 video streams. By processing the extracted motion vector fields that are directly extracted from MPEG-1 video streams in the compressed domain, through our proposed system, in order to reduce the noise within the motion vector content, obtain more robust object information, and refine this information. As a result, the object detection algorithm is more capable of accurately detecting objects with more efficient performance.