Geometric feature statistics histogram for both real-valued and binary feature representations of 3D local shape

作者：

Highlights：

• We propose a real-valued 3D local feature descriptor named GFSH.

• The reasonable construction algorithm develops the extensibility and flexibility of GFSH.

• We seamlessly extend the real-valued GFSH descriptor to the binary representation named B-GFSH.

• Across-dataset experiments and comparisons validate the superiority of GFSH and B-GFSH.

摘要

•We propose a real-valued 3D local feature descriptor named GFSH.•The reasonable construction algorithm develops the extensibility and flexibility of GFSH.•We seamlessly extend the real-valued GFSH descriptor to the binary representation named B-GFSH.•Across-dataset experiments and comparisons validate the superiority of GFSH and B-GFSH.

论文关键词：3D point cloud,Local feature descriptor,Multiple geometric features,Binary extension,Feature matching

论文评审过程：Received 12 September 2021, Revised 16 November 2021, Accepted 17 November 2021, Available online 20 November 2021, Version of Record 6 December 2021.

论文官网地址：https://doi.org/10.1016/j.imavis.2021.104339