Yet Another Computer Vision Index To Datasets (YACVID) - Details

Stand: 2017-11-22 000000m 06:17:52 - Overview

ok41
Attribute Current Content New
Name (Institute + Shorttitle)SydneyHouse HouseCraft 
Description (include details on usage, files and paper references)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.

The original SydneyHouse dataset contains 174 random houses in Sydney with:

Annotated floorplan
Adress and map
Up to three streetview images, with computed semantic features
Accurate house location and vertical heights (in accordance with streetview observations)


Hang Chu, Shenlong Wang, Raquel Urtasun, Sanja Fidler. HouseCraft: Building Houses from Rental Ads and Street Views. ECCV 2016.
 
URL Linkhttp://www.cs.toronto.edu/housecraft/ 
Files (#)174 
References (SKIPPED)
Category (SKIPPED) 
Tags (single words, spaced)house urban building city floorplan streetview semantic segmentation localization registration 
Last Changed2017-11-22 
Turing (2.12+3.25=?) :-)