Fast covariant VLAD for image search

• Main Authors: ZHAO, Wan-Lei, NGO, Chong-wah, WANG, Hanzi
• Format: text
• Language: English
• Published: Institutional Knowledge at Singapore Management University 2016
• Subjects: Circular matching; covariant pooling; covariant vector of locally aggregated descriptor (CVLAD); similar image search; Computer Sciences; Graphics and Human Computer Interfaces
• Online Access: https://ink.library.smu.edu.sg/sis_research/6310; https://ink.library.smu.edu.sg/context/sis_research/article/7313/viewcontent/07499824.pdf
• Institution: Singapore Management University

Vector of locally aggregated descriptor (VLAD) is a popular image encoding approach for its simplicity and better scalability over conventional bag-of-visual-word approach. In order to enhance its distinctiveness and geometric invariance, covariant VLAD (CVLAD) is proposed to pool local features bas...