Ambient Point Clouds for View Interpolation

  • Michael Goesele
  • Jens Ackermann
  • Simon Fuhrmann
  • Carsten Haunbold
  • Ronny Klowsky
  • Drew Steedly

Published by Association for Computing Machinery, Inc.

View interpolation and image-based rendering algorithms often produce visual artifacts in regions where the 3D scene geometry is erroneous, uncertain, or incomplete. We introduce ambient point clouds constructed from colored pixels with uncertain depth, which help reduce these artifacts while providing non-photorealistic background coloring and emphasizing reconstructed 3D geometry. Ambient point clouds are created by randomly sampling colored points along the viewing rays associated with uncertain pixels. Our realtime rendering system combines these with more traditional rigid 3D point clouds and colored surface meshes obtained using multiview stereo. Our resulting system can handle larger-range view transitions with fewer visible artifacts than previous approaches.