Automated location of SBB infrastructure objects

The aim of this project is to make a feasibility study of an automated method of locating Swiss Federal Railways infrastructure and building objects with BIM type data.

Swiss Federal Railways has an inventory database of objects that constitute the technical equipment associated with transport infrastructure and interfaces for travelers. During inspection or maintenance work, the teams working on the site must be able to unambiguously associate field objects with the elements of the database. For this, it is necessary to identify and locate an object in its context by a simple and effective method. The preferred method of location is to take a photo of the object and its context with a smartphone. On the basis of the information contained in the images, it will be sought to associate the object of interest with its counterpart stored in a database.

The purpose of this preliminary study is to evaluate the feasibility of an automated method for locating objects from Swiss Federal Railways infrastructure and buildings. It mainly consists on reviewing the state of the art on the algorithms that allow to match images of objects with a content of a geographic BIM (Building information modeling) type database.

The project is carried out by the Laboratory of Topometry (TOPO), led by Professor Bertrand Merminod, and the Computer Vision Laboratory (CVLAB), led by Professor Pascal Fua. It is financed by Swiss Federal Railways.

Principal Investigators Pierre-Yves Gillieron, Mathieu Salzmann
Period 2017-2018
Laboratories TOPO, CVLAB
Collaboration TRACE