Image and Video Retargetting by Darting

This paper considers the problem of altering an image by imperceptibly adding or removing pixels, for example, to fit a differently shaped frame with minimal loss of interesting content. We show how to construct a family of convex programs that suitably rearrange pixels while minimizing image artifacts and distortions. We call this ¨darting¨on analogy to a tailor´s dartssmall edits are discreetly distributed throughout the fabric of the image. We develop a reduction to integer dynamic programming on edit trellises, yielding fast algorithms. One- and two-pass variants of the method have 0(1) per-pixel complexity. Of the many edits that darting supports, five are demonstrated here: image retargeting to smaller aspect ratios: adding or moving or removing scene objects while preserving image dimensions: image expansion with gaps filled by a rudimentary form of texture synthesis; temporal video summarization by ¨packing ¨motion in time; and an extension to spatial video retargettin...

• Type: Conference
• Year: 2009
• Where: ICIAR
• Authors: Matthew Brand
• Keywords: Electric Research Laboratories, ICIAR 2009, Image, Mitsubishi Electric Research