A dynamic keypoint selection network for 6DoF pose estimation

作者：

Highlights：

• A simple but effective dynamic keypoint selection algorithm that leverages texture and geometry information of object.

• A filtering background points algorithm significantly improves pose estimation time efficiency.

• State-of-the-art 6DoF pose estimation performance on the YCB-Video and LineMOD datasets.

摘要

•A simple but effective dynamic keypoint selection algorithm that leverages texture and geometry information of object.•A filtering background points algorithm significantly improves pose estimation time efficiency.•State-of-the-art 6DoF pose estimation performance on the YCB-Video and LineMOD datasets.

论文关键词：6 DoF pose estimation,Dynamic keypoint selection,Scene feature fusion

论文评审过程：Received 31 October 2021, Revised 29 December 2021, Accepted 1 January 2022, Available online 5 January 2022, Version of Record 12 January 2022.

论文官网地址：https://doi.org/10.1016/j.imavis.2022.104372