In theory, it is simple to model a wall from a point cloud. You just fit a plane to the data, and you’re done. In practice, walls are much more complex entities. They are interspersed with windows and doorways, parts may be occluded by furniture and other objects, and items like bookshelves, clocks, and pictures may be attached or adjacent to the walls. In this research, we are investigating methods to more accurately model walls by explicitly reasoning about occlusions and their impact on the wall model. As part of this research, we are also developing methods to automatically recognize and model windows and doorways, even when they are partially occluded.
This project is funded in part, by the National Science Foundation (CMMI-0856558).
current staff
current contact
past head
- Antonio Adan Oliver
past staff
- Emiliano Perez Hernandez
- Enrique Valero Rodriguez