Multi-View Stereo for Community Photo Collections

We present a multi-view stereo algorithm that addresses the extreme changes in lighting, scale, clutter, and other effects in large online community photo collections. Our idea is to intelligently choose images to match, both at a per-view and per-pixel level. We show that such adaptive view selection enables robust performance even with dramatic appearance variability. The stereo matching technique takes as input sparse 3D points reconstructed from structure-from-motion methods and iteratively grows surfaces from these points. Optimizing for surface normals within a photoconsistency measure significantly improves the matching results. While the focus of our approach is to estimate high-quality depth maps, we…

• NS
  National Science Foundation
• AV
  Alexander von Humboldt-Stiftung
• UO
  University of Washington
• CN
  Centre National de la Recherche Scientifique