Real-time image segmentation on a GPU

  • Authors:

  • Alexey Abramov;Tomas Kulvicius;Florentin Wörgötter;Babette Dellen

  • Affiliations:

  • Georg-August University, Bernstein Center for Computational Neuroscience, Department for Computational Neuroscience, III Physikalisches Institut, Göttingen, Germany;Georg-August University, Bernstein Center for Computational Neuroscience, Department for Computational Neuroscience, III Physikalisches Institut, Göttingen, Germany and Department of Informat ...;Georg-August University, Bernstein Center for Computational Neuroscience, Department for Computational Neuroscience, III Physikalisches Institut, Göttingen, Germany;Bernstein Center for Computational Neuroscience, Max-Planck-Institute for Dynamics and Self-Organization, Göttingen, Germany and Institut de Robòtica i Informàtica Industrial, CSIC, ...

  • Venue:
  • Facing the multicore-challenge
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

Efficient segmentation of color images is important for many applications in computer vision. Non-parametric solutions are required in situations where little or no prior knowledge about the data is available. In this paper, we present a novel parallel image segmentation algorithm which segments images in real-time in a non-parametric way. The algorithm finds the equilibrium states of a Potts model in the superparamagnetic phase of the system. Our method maps perfectly onto the Graphics Processing Unit (GPU) architecture and has been implemented using the framework NVIDIA Compute Unified Device Architecture (CUDA). For images of 256 × 320 pixels we obtained a frame rate of 30 Hz that demonstrates the applicability of the algorithm to video-processing tasks in real-time.