Statistical models for images: Compression, restoration and synthesis

E P Simoncelli

Published in Proc 31st Asilomar Conf on Signals, Systems and Computers, vol.1 pp. 673--678, Nov 1997.
© IEEE Signal Processing Society

DOI: 10.1109/ACSSC.1997.680530

Download:

  • Reprint (pdf)
  • Reprint (ps)

  • We present a parametric statistical model for visual images in the wavelet transform domain. We characterize the joint densities of coefficient magnitudes at adjacent spatial locations, adjacent orientations, and adjacent spatial scales. The model accounts for the statistics of a wide variety of visual images. As a demonstration of this, we've used the model to design a progressive image encoder with state-of-the-art rate-distortion performance. We also show promising examples of image restoration and texture synthesis.
  • Statistical Modeling: Lyu08c, Simoncelli05a, Wainwright00, Wainwright99b
  • Compression: Buccigrossi97
  • Denoising: Lyu08, Colon08a, Portilla03
  • Texture Modeling: Portilla99
  • Neural Representation: Wainwright00c, Schwartz01
  • Listing of all publications