FAST GEOMETRIC RE-RANKING FOR IMAGE-BASED RETRIEVAL

被引：36

作者：

Tsai, Sam S. ^{[1
]}

Chen, David ^{[1
]}

Takacs, Gabriel ^{[1
]}

Chandrasekhar, Vijay ^{[1
]}

Vedantham, Ramakrishna ^{[2
]}

Grzeszczuk, Radek ^{[2
]}

Girod, Bernd ^{[1
]}

机构：

[1] Stanford Univ, Informat Syst Lab, Stanford, CA 94305 USA

[2] Nokia Res Ctr, Palo Alto, CA 94304 USA

来源：

2010 IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING | 2010年

关键词：

image-based retrieval; mobile visual search; robust features; geometric verification;

D O I：

10.1109/ICIP.2010.5648942

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

We present a fast and efficient geometric re-ranking method that can be incorporated in a feature based image-based retrieval system that utilizes a Vocabulary Tree (VT). We form feature pairs by comparing descriptor classification paths in the VT and calculate geometric similarity score of these pairs. We propose a location geometric similarity scoring method that is invariant to rotation, scale, and translation, and can be easily incorporated in mobile visual search and augmented reality systems. We compare the performance of the location geometric scoring scheme to orientation and scale geometric scoring schemes. We show in our experiments that re-ranking schemes can substantially improve recognition accuracy. We can also reduce the worst case server latency up to 1 sec and still improve the recognition performance.

引用

页码：1029 / 1032

页数：4