Speaker: Matthias Adorjan (ICGA)
Besides using high-cost laser scanning equipment to capture large point clouds for topographical or architectural purposes, nowadays other, more affordable approaches exist. Structure-from-motion (SfM) is such a low-cost photogrammetric method used to generate large point datasets. It refers to the process of estimating three-dimensional structures out of two-dimensional image sequences. These sequences can even be captured with conventional consumer-grade digital cameras.
In our work, we aim to establish a free, fully accessible and editable database for SfM data, based on the idea of the collaborative OpenStreetMap project, called OpenSfM. We are developing a web application which lets the user browse and edit existing data or upload new data in form of pictures with its corresponding camera parameters. This information is evaluated by a state-of-the-art SfM engine on the server side, which generates point clouds that are stored in the underlying database. These point clouds can be explored and downloaded with the help of a web client.
Additionally we compare different existing structure-from-motion engines with respect to their accuracy and performance.