3D-FPFH-Int: A Hybrid Geometric–Radiometric Descriptor for Anomaly Detection in Tropical Heritage Structures
Abstract
This repository contains the full implementation of the 3D-FPFH-Int descriptor, ablation experiments (geometric-only, radiometric-only, hybrid), multi-detector evaluation scripts (PatchCore, Isolation Forest, kNN), robustness testing utilities, and synthetic dataset (500 annotated point clouds). It supports reproducibility of the study “A Multi-Detector, Ablation-Controlled, and Statistically Validated Evaluation of a Hybrid 3D Geometric–Radiometric Descriptor for Structural Anomaly Detection
Fields of Study & Subjects
3D Surveying and Cultural Heritage
Infrastructure Maintenance and Monitoring
Anomaly Detection Techniques and Applications
Publication details
- Journal / source
- Zenodo (CERN European Organization for Nuclear Research)
- Publication date
- 3 March 2026
- Language
- EN
- Document type
- Dataset