A repeatable and robust local reference frame for 3D surface matching

作者：

Highlights：

• An effective local reference frame (LRF) based on feature transformation is proposed for 3D surface matching. The proposed LRF achieves significant higher repeatability and stronger robustness.

• A scale strategy is proposed to improve the robustness of the LRF with respect to mesh decimation. The technique is also applicable to other LRFs.

• Experimental results demonstrate that our method achieves significant higher repeatability and stronger robustness than the state-of-the-arts under Gaussian noise, shot noise and mesh resolution variation.

摘要

•An effective local reference frame (LRF) based on feature transformation is proposed for 3D surface matching. The proposed LRF achieves significant higher repeatability and stronger robustness.•A scale strategy is proposed to improve the robustness of the LRF with respect to mesh decimation. The technique is also applicable to other LRFs.•Experimental results demonstrate that our method achieves significant higher repeatability and stronger robustness than the state-of-the-arts under Gaussian noise, shot noise and mesh resolution variation.

论文关键词：Local reference frame,3D Surface description,Feature transformation,Local characteristics,Scale strategy

论文评审过程：Received 29 April 2019, Revised 8 December 2019, Accepted 24 December 2019, Available online 25 December 2019, Version of Record 29 December 2019.

论文官网地址：https://doi.org/10.1016/j.patcog.2019.107186