Fisher encoding of differential fast point feature histograms for partial 3D object retrieval. (July 2016)

• Record Type: Journal Article
• Authors: Savelonas, Michalis A.; Pratikakis, Ioannis; Sfikas, Konstantinos
• Is Part Of: Pattern recognition. Volume 55(2016:Jul.)
• Page Start: 114
• Page End: 124
• Publication Date: 2016-07
• Subjects: 3D object retrieval -- Partial matching -- Local descriptors -- Fisher encoding; Pattern perception -- Periodicals; Perception des structures -- Périodiques; Patroonherkenning; 006.4
• DOI: 10.1016/j.patcog.2016.02.003
• Languages: English
• ISSNs: 0031-3203

Abstract

Abstract: Partial 3D object retrieval has attracted intense research efforts due to its potential for a wide range of applications, such as 3D object repair and predictive digitization. This work introduces a partial 3D object retrieval method, applicable on both point clouds and structured 3D models, which is based on a shape matching scheme combining local shape descriptors with their Fisher encodings. Experiments on the SHREC 2013 large-scale benchmark dataset for partial object retrieval, as well as on the publicly available Hampson pottery dataset, demonstrate that the proposed method outperforms seven recently evaluated partial retrieval methods. Abstract : Highlights: First application of Fisher encoding in 3D object retrieval. The dFPFH extends FPFH, capturing more accurately local geometric transitions. The proposed similarity addresses partiality. The proposed method outperforms state-of-the-art in SHREC 2013 dataset. The proposed method outperforms state-of-the-art in cultural heritage datasets.