Image-based window detection: an overview
Automated segmentation of buildings’ façade and detection of its elements is of high relevance in various fields of research as it, e. g., reduces the effort of 3 D reconstructing existing buildings and even entire cities or may be used for navigation and localization tasks. In recent years, several...
| Main Authors: | , , |
|---|---|
| Format: | Conference or Workshop Item |
| Published: |
2016
|
| Online Access: | https://eprints.nottingham.ac.uk/34763/ |
| Summary: | Automated segmentation of buildings’ façade and detection of its elements is of high relevance in various fields of research as it, e. g., reduces the effort of 3 D reconstructing existing buildings and even entire cities or may be used for navigation and localization tasks. In recent years, several approaches were made concerning this issue. These can be mainly classified by their input data which are either images or 3 D point clouds. This paper provides a survey of image-based approaches. Particularly, this paper focuses on window detection and therefore groups related papers into the three major detection strategies. We juxtapose grammar based methods, pattern recognition and machine learning and contrast them referring to their generality of application. As we found out machine learning approaches seem most promising for window detection on generic façades and thus we will pursue these in future work. |
|---|