Ponte San Lorenzo, a case study for the comparison of image-based survey tools. NeRF as an alternative to photogrammetry

Maurizio Perticarini, Andrea Giordano

Abstract

The aim of the research is to experiment with the new possibilities offered by artificial intelligence applied to surveying. Ponte San Lorenzo, a perfectly preserved archaeological artefact under Via Riviera dei Ponti Romani in Padua, one of the case studies of the 2018 project of the University of Padua entitled: PD-Invisible: PaDova INnovative VISions - visualizations and Imaginings Behind the city Learning, was the object of an accurate integrated image and range-based survey and made it possible to compare the different 3D acquisition methodologies of the cultural heritage. The Bridge is now digitally rendered through the new system developed by Nvidia called Instant Nerf and the new web-based software Luma AI which exploit machine learning and Volume Ray Merching for the creation of virtual twins. The same database of photographs taken for photogrammetry was used as a training database for the algorithm. The research aims to evaluate the results of the new survey system and to compare them with the results of the methods used in the past.

Download

IMEKO-MetroArchaeo-2023-152.pdf

DOI

10.21014/tc4-ARC-2023.152

IMEKO TC

TC4 - Measurement of Electrical Quantities

Back to the proceedings