Detection and classification of objects with machine/deep learning for asset management in buildings

M-GEO
ACQUAL
Staff Involved
Additional Remarks

Students should have suitable programming skills (Python/c++). The research is done in collaboration with a company, CGI. Internship options can be discussed.

Topic description

Asset management is important for commercial and public buildings. One example is hospitals, which contain a lot of medical equipment but also small safety assets such as fire detectors, sprinklers, and extinguishers. In order to digitally inspect the presence and condition of these assets, indoor mobile mapping system data needs to be automatically processed. The processing methods consist of the following problems applied onto RGB image data and 3D point clouds: object detection, object classification, object localization. This MSc topic is done jointly with a Dutch company, CGI. Internship option is available.

Topic objectives and methodology

Machine/deep learning approaches for object detection, object classification, and object localization methods applied mainly on RGB images (and optionally on 3D point clouds)