Google Street View provides visual views of cities and other locations. Though they provide important information, images are not updated very regularly. This is because of the prohibitive cost required to take pictures continuously. Google would need to run their fleet camera mounted vans 24x7x365 all around the country to gather images which would be prohibitively expensive. To solve this temporal inconsistency problem, we have developed a system that takes advantage of social networking to gather and position images. Since the majority of mobile phones are camera, GPS, and accelerometer enabled, a social network of people using these phones can gather geo-tagged images in a timely manner. These images then can be stored on a back-end server. The problem with using mobile phone camera is image quality. However, various sophisticated image stitching and image filtering algorithms are developed to enhance the quality of the image. Also, on the other hand after the geo-tagged images have been enhanced they can be stitched together to form huge panoramic views of group of images resulting into a better view of the scene. Later the panoramic view can be overlaid on Google street view to provide non stale view of the world around us.
展开▼
