Seam carving for content-aware image resizing

Abstract

Figure 1: A seam is a connected path of low energy pixels in an image. On the left is the original image with one horizontal and one vertical seam. In the middle the energy function used in this example is shown (the magnitude of the gradient), along with the vertical and horizontal path maps used to calculate the seams. By automatically carving out seams to reduce image size, and inserting seams to extend it, we achieve content-aware resizing. The example on the top right shows our result of extending in one dimension and reducing in the other, compared to standard scaling on the bottom right. AbstractEffective resizing of images should not only use geometric constraints, but consider the image content as well. We present a simple image operator called seam carving that supports content-aware image resizing for both reduction and expansion. A seam is an optimal 8-connected path of pixels on a single image from top to bottom, or left to right, where optimality is defined by an image energy function. By repeatedly carving out or inserting seams in one direction we can change the aspect ratio of an image. By applying these operators in both directions we can retarget the image to a new size. The selection and order of seams protect the content of the image, as defined by the energy function. Seam carving can also be used for image content enhancement and object removal. We support various visual saliency measures for defining the energy of an image, and can also include user input to guide the process. By storing the order of seams in an image we create multi-size images, that are able to continuously change in real time to fit a given size.

Improved Seam Merging for Content-Aware Image Resizing

“…In image enlargement, seam carving once removes seams, and then duplicates corresponding seams on an original image. The reason is straightforward pixel inserting creates a stretching artifact [3]. Additionally, seam carving needs to break the pro- cess into several steps in the case of excessive image enlarging (for instance, greater than 50%).…”

“…Pixel importance can be measured by intensity gradient, saliency measure, object detections, user inputs, and many other criteria [3]. The determination of which criteria should be used as image importance depends on context.…”

“…For effective image resizing, several content-aware image resizing methods have been proposed [1], [2]. Seam carving [3] is one of the approaches for content-aware image resizing. It repeatedly removes an 8-connected path of pixels, called seam, from top to bottom or from left to right.…”

Content-Aware Convolutional Neural Network for Object Recognition Task

“…Seam carving is an algorithm for content-aware image resizing [5], introduced by Shai Avidan and Ariel Shamir. Unlike other regular image resizing technique, seam carving is able to resize image while maintaining its important features intact.…”

“…Seam carving is a computer vision approach to resize an image while maintaining the important object/pixels inside. This can be done by removing the least noticeable pixels in the image [5]. Assume that a pixel is similar to the surrounding pixels, then removing it may be unnoticed.…”