Chapter 32 — 3-D Vision for Navigation and Grasping

Finding paths through the world's photos

When a scene is photographed many times by different people, the viewpoints often cluster along certain paths. These paths are largely specific to the scene being photographed and follow interesting patterns and viewpoints. We seek to discover a range of such paths and turn them into controls for image-based rendering. Our approach takes as input a large set of community or personal photos, reconstructs camera viewpoints, and automatically computes orbits, panoramas, canonical views, and optimal paths between views. The scene can then be interactively browsed in 3-D using these controls or with six DOF free-viewpoint control. As the user browses the scene, nearby views are continuously selected and transformed, using control-adaptive reprojection techniques.
Noah Snavely, Rahul Garg, Steven M. Seitz, Richard Szeliski
Special thanks to Kevin Chiu and Andy Hou for their help with this project. We also appreciate the many Flickr users who allowed us to use their photos.