The goal of the CITYFIT project is, given highly redundant input imagery and range maps from an arbitrary building in Graz, to synthesize a shape grammar that, when evaluated, creates a clean, CAD-quality reconstruction of that building that fits the original data very closely and makes the semantics of all major architectural features explicit. These shape semantics can even be transferred back to inform the original data, so each of these "semantically enriched" data points can tell whether it belongs to ground, wall, or door. This information is used to create detailed façade models.
- ICG: Hayko Riemenschneider, Horst Bischof
- CGV: Ulrich Krispel, Wolfgang Thaller, Sven Havemann,Dieter Fellner
- Vexcel: Michael Grabner, Konrad Karner
CityFit - Urban environment reconstruction
Large scale urban environment parsing is a challenging topic with the goal to extract semantic connected structures based on 3D layout, semantic categories and object detection.
- Analysis, segmentation, and classification of 3D pointclouds
- Piecewise planar modeling of building façades
- Joining detection and segmentation for accurate object localization
- Multi-structure fitting for detections and parametric models
- Symmetry and repetition analysis for high-level structure detection
Example high-quality knowledge of façades including architectural elements such as windows, door, balconies, ledges, roof lines, and piecewise planar façade surface:
Overview of some of the modeled façades in Graz: