In HouseCraft, we utilize rental ads to create realistic textured 3D models of building exteriors. In particular, we exploit the address of the property and its floorplan, which are typically available in the ad. The address allows us to extract Google StreetView images around the building, while the building’s floorplan allows for an efficient parametrization of the building in 3D via a small set of random variables. Our approach is able to precisely estimate the geometry and location of the property, and can create realistic 3D building models.
HouseCraft covered by 2 Minute Papers. Thank you Károly!
Starting from the address of the house (a), we propose an energy minimization framework which jointly reasons about the height of each floor, the vertical positions of windows and doors, as well as the precise location of the building in the world’s map, by exploiting several geometric and semantic cues from the StreetView imagery (b-f).
Paper and supplemental material:
Demo and results:
demo_eccv16 from Hang Chu on Vimeo.
174 random houses in Sydney with:
An extra dataset derived from SydneyHouse, that contains rectified house facade images with labelled window, door, and garage.
An extra dataset derived from SydneyHouse, that contains dense point correspondences between streetviews. Currently 17.3K points, with sample rate 1 point/m.
A WebGL-based annotation tool we used to create house ground truth information.