Computer Graphics
TU Braunschweig

Efficient GPU Based Sampling for Scene-Space Video Processing

Efficient GPU Based Sampling for Scene-Space Video Processing

We describe a method to efficiently collect and filter a large set of 2D pixel observations of unstructured 3D

points, with applications to

scene-space aware

video processing. One of the main challenges in scene-space video

processing is to achieve reasonable computation time despite the very large volumes of data, often in the order of

billions of pixels. The bottleneck is determining a suitable set of candidate samples used to compute each output

video pixel color. These samples are observations of the same 3D point, and must be gathered from a large number

of candidate pixels, by volumetric 3D queries in scene-space. Our approach takes advantage of the spatial and

temporal continuity inherent to video to greatly reduce the candidate set of samples by solving 3D volumetric

queries directly on a series of 2D projections, using out-of-core data streaming and an efficient GPU producer-

consumer scheme that maximizes hardware utilization by exploiting memory locality. Our system is capable of

processing over a trillion pixel samples, enabling various scene-space video processing applications on full HD

video output with hundreds of frames and processing times in the order of a few minutes.

Author(s):Felix Klose, Oliver Wang, Jean-Charles Bazin, Alexander Sorkine-Hornung, Marcus Magnor
Published:October 2015
Type:Article in conference proceedings
Book:Proc. Vision, Modeling and Visualization (VMV)
Presented at:Vision, Modeling and Visualization (VMV)
Project(s): Scene-Space Video Processing  Reality CG 

  title = {Efficient {GPU} Based Sampling for Scene-Space Video Processing},
  author = {Klose, Felix and Wang, Oliver and Bazin, Jean-Charles and Sorkine-Hornung, Alexander and Magnor, Marcus},
  booktitle = {Proc. Vision, Modeling and Visualization ({VMV})},
  organization = {Eurographics},
  editor = {D. Bommes, T. Ritschel, and T. Schultz},
  pages = {103--110},
  month = {Oct},
  year = {2015}