Exploiting semantic knowledge for robot object recognition

作者：

Highlights：

• Semantic knowledge is exploited to train a Probabilistic Graphical Model.

• This knowledge is gathered through expert/human elicitation.

• The recognition system performs in scenes perceived by a mobile robot.

• Validation is conducted with 3d-point cloud datasets.

• A comparison with other state-of-the-art methods is carried out.

摘要

•Semantic knowledge is exploited to train a Probabilistic Graphical Model.•This knowledge is gathered through expert/human elicitation.•The recognition system performs in scenes perceived by a mobile robot.•Validation is conducted with 3d-point cloud datasets.•A comparison with other state-of-the-art methods is carried out.

论文关键词：Semantic knowledge,Human elicitation,Object recognition,Probabilistic Graphical Models,Autonomous robots

论文评审过程：Received 7 January 2015, Revised 13 May 2015, Accepted 27 May 2015, Available online 8 June 2015, Version of Record 31 July 2015.

论文官网地址：https://doi.org/10.1016/j.knosys.2015.05.032