Locating Façades using Single Perspective Images, Cube Maps and Inaccurate GPS Coordinates

Search for information about monuments, business establishments and façades is a recurring task for travelers and citizens of large cities. However, even when the person is in front of the place of interest, the language and the inaccuracy of the GPS receivers integrated in mobile phones may make it difficult the precise identification of the site. We present a tool that is capable to identify the correct geographic coordinates of the place of interest. Given a picture of the place and inaccurate geographic coordinates provided by a smartphone, the building is located in a web mapping service (e.g., Google Maps) by looking for the image of its façade into panoramic views of the streets (e.g., Google Street View) nearby the given geographic coordinates. We demonstrate the proposed approach and algorithms by building a prototype of the application. Our solution employs graph traversal algorithms, computer vision and image processing techniques such as ASIFT, panoramic projection, triangulation, and histogram analysis. The effectiveness of our solution was asserted by automatically retrieving the geographic coordinates of façades in our hometown. Keywords-façade; ASIFT; mobile device; virtual map; visual identification; GPS coordinates.