PHOTOGRAMMETRY WITH LIDAR IPAD PRO

Abstract

Digital documentation of cultural heritage is a field that has grown rapidly in the last few years. The technological advancements in computing and the increased collaboration between fields keep offering researchers new tools with which they can capture objects, monuments or even entire archaeological sites, thus permitting non-destructive visual and geometric research and reproduction. Due to this rapid pace at which technology is advancing, such tools are now also available to a much wider audience, through everyday handheld devices. Smartphone and tablet cameras have improved significantly, to the point where they can work very effectively in place of a Digital Single Lens Reflex (DSLR) camera, in a photogrammetry setup. Accurate 3D reconstruction of cultural heritage objects with a smartphone and a computer is now possible and accessible to a multitude of people. Students and educators can capture objects in digital format, which they can then use for research or presentations purposes, non-destructive experimentation or even working 3D models. Publishing houses and content creators that design educational content or serious games, can also take advantage of a fast and easy way to create accurate 3D models of real-world objects and incorporate them in their work. Some of the newer handheld devices, mainly Apple’s iPhones and iPads, have incorporated Light Detection and Ranging (LiDAR) sensors, for use with Augmented and Virtual Reality applications. These sensors can also offer a great degree of accuracy when used for photogrammetry. In this study we compare the photogrammetric process as performed by the 2021 iPad Pro with the M1 processor, in situ, to a DSLR camera and PC setup. We test the speed of capturing and processing in each case, and we also directly compare the 3D models’ points and faces, as a way to measure the detail present in the final output. We use the iPadOS Polycam application, which exploits the device’s built-in LiDAR sensor, to perform in situ photogrammetry. At the same time, when capturing is complete on the iPad, and to avoid lighting variations between processes, we use a DSLR camera for photography, that is similar to the iPad’s camera in Megapixel. We then have a recent PC with Agisoft’s Metashape perform the off-site photogrammetric process. The results indicate that the M1 iPad can be very fast in processing and with a very high degree of accuracy, thus making it a viable replacement for a camera and PC setup, and a valuable tool that offers accurate mobile photogrammetry, while also providing a solution for budgetary constraints.