Image-based 3D reconstruction of bridges
Table of contents
Project data
| Titel | Title TP der TU Dresden im Verbundprojekt Pic2Bridge: Bildbasiertes Ingenieurdesign für Brücken | SP of TU Dresden in the joint research project Pic2Bridge: Image-based engineering design for bridges Förderer | Funding Bundesministerium für Digitales und Verkehr (BMDV) / mFUND Zeitraum | Period 08/2024 – 07/2025 Verbundvorhabenleiter | Leader of joint research project Prof. Dr.-Ing. Steffen Marx Teilprojektleiter | Subproject manager Prof. Dr.-Ing. Steffen Marx Bearbeiterin | Contributor Samar Aqlan, M.Sc. |
Short description
Bridge design is influenced by the experience and expertise of the designing engineer, which can lead to significant differences in quality. Efficient methods for analyzing existing bridge structures based on open available data sources, such as image material, do not currently exist. The development of innovative methods is necessary to systematically extract information from images and prepare it for the design process.
Pic2Bridge aims to investigate the feasibility of 3D reconstruction of bridges using a limited number of images. Data from open sources, such as the Mobilithek, the Deutsche Fotothek, and Structurae, will be integrated as inputs for 3D modeling. The approach seeks to expand the possibilities of data acquisition and processing, thus enabling precise and comprehensive reconstructions of bridge structures. The project begins with the systematic collection of data from existing open sources and its storage in a structured image database. This is followed by the annotation and further processing of the data, while AI methods for extracting image information are developed. An AI-based implementation is intended to support the automation of 3D reconstruction and optimize existing methods for bridge design.