Share Email Print
cover

Proceedings Paper

SORB: improve ORB feature matching by semantic segmentation
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

Feature matching is at the base of many computer vision algorithms such as SLAM, which is a technology widely used in the area from intelligent vehicles (IV) to assistance for the visually impaired (VI). This article presents an improved detector and a novel semantic-visual descriptor, coined SORB (Semantic ORB), combining binary semantic labels and traditional ORB descriptor. Compared to the original ORB feature, the new SORB performs better in uniformity of distribution and accuracy of matching. We demonstrate it through experiments on some open source datasets and several real-world images obtained by RealSense.

Paper Details

Date Published: 4 October 2018
PDF: 7 pages
Proc. SPIE 10799, Emerging Imaging and Sensing Technologies for Security and Defence III; and Unmanned Sensors, Systems, and Countermeasures, 107990Z (4 October 2018); doi: 10.1117/12.2325423
Show Author Affiliations
Hao Chen, Zhejiang Univ. (China)
Kaiwei Wang, Zhejiang Univ. (China)
Weijian Hu, Zhejiang Univ. (China)
Lei Fei, Zhejiang Univ. (China)


Published in SPIE Proceedings Vol. 10799:
Emerging Imaging and Sensing Technologies for Security and Defence III; and Unmanned Sensors, Systems, and Countermeasures
Gerald S. Buller; Markus Mueller; Richard C. Hollins; Robert A. Lamb, Editor(s)

© SPIE. Terms of Use
Back to Top