Global image representation using Locality-constrained Linear Coding for large-scale image retrieval

Yu Hsing Wu, Wei Lin Ku, Wen-Hsiao Peng, Hung Chun Chou

研究成果: Conference contribution同行評審

3 引文 斯高帕斯(Scopus)

摘要

This paper proposes a global image representation based on Locality-constrained Linear Coding (LLC), with an aim to simplify the encoding process of local descriptors so as to facilitate large-scale image retrieval. Starting from the state-of-the-art Fisher Vector (FV) representation, we replace the computation of sophisticated posterior probabilities with simpler LLC. We then conduct several empirical studies to investigate the effects and benefits of this change and to adapt the other terms in FV for a better trade-off between performance and complexity. The result is a simpler global descriptor that combines the merits of both FV and LLC. Experimental results show that when compared with other similar works, our scheme not only brings performance benefits in mean Average Precision, but also offer complexity advantages.

原文English
主出版物標題2014 IEEE International Symposium on Circuits and Systems, ISCAS 2014
發行者Institute of Electrical and Electronics Engineers Inc.
頁面766-769
頁數4
ISBN(列印)9781479934324
DOIs
出版狀態Published - 1 1月 2014
事件2014 IEEE International Symposium on Circuits and Systems, ISCAS 2014 - Melbourne, VIC, Australia
持續時間: 1 6月 20145 6月 2014

出版系列

名字Proceedings - IEEE International Symposium on Circuits and Systems
ISSN(列印)0271-4310

Conference

Conference2014 IEEE International Symposium on Circuits and Systems, ISCAS 2014
國家/地區Australia
城市Melbourne, VIC
期間1/06/145/06/14

指紋

深入研究「Global image representation using Locality-constrained Linear Coding for large-scale image retrieval」主題。共同形成了獨特的指紋。

引用此