An Effective Image Retrieval System in Object-region Using Family-based Spatial Consistency Filtration
Building an appropriate spatial consistency is the innermost core of improving retrieval performance. To address this problem we introduce a novel image retrieval mechanism based on the family filtration in the object-region. We elaborate on the performance of the retrieval system on a dataset of about 2000 images selected from the key frame of the “TRECVID2005. First we supply an object-region by selecting a rectangle of a query image, and the system returns a ranked listed of images that contain the same object, retrieved from the corpus based on the 100 image sets, as the result of the first-rank. To further improve query performance, we add an efficient spatial consistency stage to re-rank the results returned by the first-rank. Our experiments show that the retrieval mechanism has a vast major effect on improving the search quality.
Image retrieval Object-region Family-based spatial consistency filtration
Jing Sun Yingjie Xing
School of Mechanical Engineering Dalian University of Technology Dalian, Liaoning Province, China
国际会议
2008 Sino-European Workshop on Intelligent Robots and Systems(SEIROS08)(第一届中欧智能系统及机器人国际学术研讨会)
重庆
英文
1-6
2008-12-11（万方平台首次上网日期，不代表论文的发表时间）