Sparse coding of natural images using a prior on edge co-occurences

• Browse the technical programme of the conference

• Download the corresponding manuscript.

Titre

Sparse Coding Of Natural Images Using A Prior On Edge Co-Occurrences

Résumé

Oriented edges in images commonly occur in co-linear and co-circular

arrangements, obeying the good continuation law of Gestalt psychology. The human visual system appears to exploit this property of images, with contour detection, line completion, and grouping performance well predicted by such an association field between edge elements \citep{Field93,Geisler01}. In this paper, we show that an association field of this type can be used to enhance the sparse representation of natural images. First, we define the SparseLets framework as an efficient representation of images based on a discrete wavelet transform. Second, we extract second-order information about edge co-occurrences from a set of images of natural scenes. Finally, we incorporate this prior information into our framework and show that it allows for the extraction of features relevant to natural scenes, like a round shape. This novel approach points the way to practical computer vision algorithms with human-like performance.

Slides

slides, slides with notes

Code

For more information on Matching Pursuit on natural images, follow The hitchhiker guide to Matching Pursuit.

reference

• Laurent U. Perrinet, James A. Bednar. Sparse Coding Of Natural Images Using A Prior On Edge Co-Occurences, URL URL2 . In European Signal Processing Conference 2015 (EUSIPCO 2015), Nice, France, 2015 abstract
  Oriented edges in images of natural scenes tend to be aligned in co-linear or co-circular arrangements, with lines and smooth curves more common than other possible arrangements of edges (the "good continuation law" of Gestalt psychology). The visual system appears to take advantage of this prior knowledge about natural images, with human contour detection and grouping performance well predicted by such an "association field" between edge elements. Geisler et al (2001) have estimated this prior information available to the visual system by extracting contours from a database of natural images, and showed that these statistics could predict behavioral data from humans in a line completion task. In this paper, we show that an association field of this type can be used for the sparse representation of natural images.
  .

TagYear15 TagBrainScales TagPublicationsArticles TagTalks