Shift-Map Image Editing - Retargeting
Shift-Map Image Editing Image Retargeting
Shift-Map Home Retargeting Inpainting Reshuffling Composition More Examples
Click on an image to see it in full size

In these retargeting examples we constrained the leftmost and rightmost columns of the output image to come from the leftmost and rightmost columns of the input image.

The shift maps indicate for each pixel in the output image, its mapping (shift) into the input image.
In this example horizontal shifts varies between 0 pixels (leftmost column) and half image width (rightmost column), vertical shift varies between -15 to 15 pixels. values are stretched for display.


Input Image

Result

Horizontal Shift Map

Vertical Shift Map


This is a comparison with other retargeting methods - reducing width by 50% (see references below). We used the same set of parameters for all inputs, and allowed shift only in x-axis.


Input Image

Seam Carving

Wolf et al.

Wang et al.

Shift-Map

References:

  • S. Avidan and A. Shamir. Seam carving for content-aware image resizing. ACM Trans. Graph., 26(3):10, 2007.
  • M. Rubinstein, A. Shamir, and S. Avidan. Improved seam carving for video retargeting. In SIGGRAPH`08, pages 1-9, 2008.
  • Y. Wang, C. Tai, O. Sorkine, and T. Lee. Optimized scale and- stretch for image resizing. ACM Trans. Graph., 27(5):1-8, 2008.
  • L. Wolf, M. Guttmann, and D. Cohen-Or. Non-homogeneous content-driven video-retargeting. In ICCV`07, 2007.