HoPPF: A novel local surface descriptor for 3D object recognition
作者：
Highlights：
• A novel 3D descriptor named Histograms of Local Point Pair Features (HoPPF) is proposed.
• HoPPF outperforms the other state-of-the-arts in terms of descriptiveness, robustness, and efficiency.
• Three modules for HoPPF, including data preprocessing, local point pair set division, distribution matrix generation are proposed.
• The local-point-pair-set-division technique can be grafted to other descriptors to improve their performance.
• Accurate recognition results on real datasets validate that the HoPPF can be applied to 3D vision.
摘要
•A novel 3D descriptor named Histograms of Local Point Pair Features (HoPPF) is proposed.•HoPPF outperforms the other state-of-the-arts in terms of descriptiveness, robustness, and efficiency.•Three modules for HoPPF, including data preprocessing, local point pair set division, distribution matrix generation are proposed.•The local-point-pair-set-division technique can be grafted to other descriptors to improve their performance.•Accurate recognition results on real datasets validate that the HoPPF can be applied to 3D vision.
论文关键词：Local feature descriptor,3D representation,Feature matching,Shape retrieval,Object recognition
论文评审过程：Received 29 May 2019, Revised 30 December 2019, Accepted 11 February 2020, Available online 12 February 2020, Version of Record 15 February 2020.
论文官网地址：https://doi.org/10.1016/j.patcog.2020.107272
