Trigonometric projection statistics histograms for 3D local feature representation and shape description

• Published in: Pattern recognition Vol. 143; p. 109727
• Main Authors: Liu, Xingsheng, Li, Anhu, Sun, Jianfeng, Lu, Zhiyong
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.11.2023
• Subjects: 3D feature descriptor; Local reference frame; Object recognition; Shape registration; Trigonometric projection mechanism; Trigonometric projection mechanism; Object recognition; Local reference frame; 3D feature descriptor; Shape registration
• ISSN: 0031-3203; 1873-5142
• DOI: 10.1016/j.patcog.2023.109727

•A novel 3D local descriptor using trigonometric projection statistics histograms.•Local reference frame construction based on a multi-attribute weighting strategy.•Trigonometric projection mechanism for descriptive and robust shape representation.•Spatial-geometrical information fusion with storage compactness and time efficiency. Feature representation as a significant approach to three-dimensional (3D) shape description has been widely employed in computer vision. However, most existing methods are suffering from the emerging challenges for descriptiveness, robustness and efficiency. This paper presents a novel feature descriptor named trigonometric projection statistics histograms (TPSH). By constructing the repeatable local reference frame based on a multi-attribute weighting strategy, TPSH can address many prevailing nuisances such as noise, occlusion and varying resolution. The trigonometric projection mechanism is originally proposed for TPSH generation, which combines two perspective views to encode both spatial distribution and geometrical measurements from local shape into statistics histograms. The experimental evaluation on public datasets proves that TPSH outperforms state-of-the-art methods in descriptiveness and robustness while maintaining storage compactness and computational efficiency. It is demonstrated that TPSH can not only be suited for 3D object recognition and shape registration, but also generalized across various acquisition devices, data modalities and application scenarios.