Object recognition using local invariant features for robotic applications: A survey

作者：

Highlights：

• A complete analysis of object recognition methods based on local invariant features from a robotics perspective is given.

• A brief description of the main approaches reported in the literature is included.

• Methods are analyzed by considering the main requirements of robotics applications.

• Best performing object recognition systems are built using ORB–ORB and DoG–SIFT keypoint-descriptor combinations.

摘要

•A complete analysis of object recognition methods based on local invariant features from a robotics perspective is given.•A brief description of the main approaches reported in the literature is included.•Methods are analyzed by considering the main requirements of robotics applications.•Best performing object recognition systems are built using ORB–ORB and DoG–SIFT keypoint-descriptor combinations.

论文关键词：Local invariant features,Object recognition,Local descriptors,Local interest points

论文评审过程：Received 2 February 2016, Revised 15 March 2016, Accepted 11 May 2016, Available online 24 May 2016, Version of Record 23 June 2016.

论文官网地址：https://doi.org/10.1016/j.patcog.2016.05.021