TY - GEN

T1 - Removing pedestrians from Google Street View images

AU - Flores, Arturo

AU - Belongie, Serge

PY - 2010

Y1 - 2010

N2 - Since the introduction of Google Street View, a part of Google Maps, vehicles equipped with roof-mounted mobile cameras have methodically captured street-level images of entire cities. The worldwide Street View coverage spans over 10 countries in four different continents. This service is freely available to anyone with an internet connection. While this is seen as a valuable service, the images are taken in public spaces, so they also contain license plates, faces, and other information information deemed sensitive from a privacy standpoint. Privacy concerns have been expressed by many, in particular in European countries. As a result, Google has introduced a system that automatically blurs faces in Street View images. However, many identifiable features still remain on the un-blurred person. In this paper, we propose an automatic method to remove entire pedestrians from Street View images in urban scenes. The resulting holes are filled in with data from neighboring views. A compositing method for creating "ghost-free" mosaics is used to minimize the introduction of artifacts. This yields Street View images as if the pedestrians had never been there. We present promising results on a set of images from cities around the world.

AB - Since the introduction of Google Street View, a part of Google Maps, vehicles equipped with roof-mounted mobile cameras have methodically captured street-level images of entire cities. The worldwide Street View coverage spans over 10 countries in four different continents. This service is freely available to anyone with an internet connection. While this is seen as a valuable service, the images are taken in public spaces, so they also contain license plates, faces, and other information information deemed sensitive from a privacy standpoint. Privacy concerns have been expressed by many, in particular in European countries. As a result, Google has introduced a system that automatically blurs faces in Street View images. However, many identifiable features still remain on the un-blurred person. In this paper, we propose an automatic method to remove entire pedestrians from Street View images in urban scenes. The resulting holes are filled in with data from neighboring views. A compositing method for creating "ghost-free" mosaics is used to minimize the introduction of artifacts. This yields Street View images as if the pedestrians had never been there. We present promising results on a set of images from cities around the world.

UR - http://www.scopus.com/inward/record.url?scp=77956523472&partnerID=8YFLogxK

U2 - 10.1109/CVPRW.2010.5543255

DO - 10.1109/CVPRW.2010.5543255

M3 - Conference article

AN - SCOPUS:77956523472

SP - 53

EP - 58

JO - 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition - Workshops, CVPRW 2010

JF - 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition - Workshops, CVPRW 2010

T2 - 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition - Workshops, CVPRW 2010

Y2 - 13 June 2010 through 18 June 2010

ER -