Seam Carving for Content-Aware Image Resizing

By Shai Avidan, Ariel Shamir. 2007

1,877 citations recorded by the source corpus.

Effective 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.

Subfields: Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

Citations in the corpus

Generative Visual Manipulation on the Natural Image Manifold (2016, background): “ certain goal specied by the user. Examples of basic editing include changing the color properties of an image either globally [6] or locally [7]. More advanced editing methods such as image warping [8,9,10] or structured image editing [11] intelligently reshue the pixels in an image following user’”
SinGAN: Learning a Generative Model From a Single Natural Image (2019, neutral): “We illustrate this below through comparison to seam carving [1].”
Digits: freehand 3D interactions anywhere using a wrist-worn gloveless sensor (2012, methodology): “A technique utilized in seam carving [2] is used to disambiguate the main vertical boundary between pairs of ﬁngers in the IR illuminated image.”
Segmenting Salient Objects from Images and Videos (2010, background): “However, reliable saliency detection methods would be useful in many applications like adaptive compression and scaling [2, 3], unsupervised image segmentation [4, 5], and object recognition [6, 7].”
Differentiable vector graphics rasterization for editing and learning (2020, methodology): “To illustrate this, we have implemented a vector retargeting operator that changes the aspect ratio of a vector graphic using the seam carving algorithm in the raster domain [Avidan and Shamir 2007].”
Deep Level Sets for Salient Object Detection (2017, methodology): “Inspired by this biological capability, visual saliency computation is introduced into the field of computer vision for its potential to enhance tasks like image processing [4, 12], image understanding [58, 63], video analysis and compression [13, 45].”
What Do Different Evaluation Metrics Tell Us About Saliency Models? (2016, background): “[54], [87], thumbnailing [55], [76], content-aware image re-targeting and cropping [3], [4], [69], [72], [85], rendering and visualization [42], [52], collage [31], [86] and artistic rendering [20], [41] require ranking (by importance or saliency) different image regions.”
Deep Saliency with Encoded Low Level Distance Map and High Level Features (2016, background): “This topic has received a great deal of attention in computer vision and cognitive science because of its wide range of applications such as content-aware image cropping [22] and resizing [3], video summarization [24], object detection [20], and person re-identiﬁcation [31].”
A system for retargeting of streaming video (2009, background): “A different class of approaches removes unimportant content from the interior of the images or video [Avidan and Shamir 2007; Rubinstein et al. 2008].”
Visual Saliency Detection Based on Multiscale Deep CNN Features (2016, background): “Such tasks include image cropping [6], retargeting [7], summarization [8] and thumbnail generation [9].”
Context-aware saliency detection (2010, background): “, [1, 21, 17, 20]) that make use of these regions have led to different definitions and interesting detection algorithms.”
Digital image forensics: a booklet for beginners (2010, methodology): “Initially proposed for content-aware image resizing [5], the algorithm is based on the”
Editing Conditional Radiance Fields (2021, background): “There has been much work on interactive tools for selecting and cloning regions [2, 35, 56, 60] and editing single still images [3, 6, 7, 34].”
SuperCNN: A Superpixelwise Convolutional Neural Network for Salient Object Detection (2015, methodology): “The comparison is conducted between the proposed method and the context-aware saliency detection method CA (Goferman et al. 2010), using the non-homogeneous resizing technique proposed by Avidan and Shamir (2007).”
Attentive Systems: A Survey (2017, methodology): “For example, attention-based recurrent networks have been successfully applied to a wide variety of tasks including handwriting synthesis (Graves 2013), machine translation (Bahdanau et al. 2014), image caption generation (Xu et al. 2015) and visual object classification (Mnih et al. 2014). Human ga”
Instance-Level Salient Object Segmentation (2017, background): “bject regions in images. It is a fundamental problem in computer vision and has served as a pre-processing step to facilitate a wide range of vision applications including content-aware image editing [4], object detection [38], and video summarization [36]. Recently the accuracy of salient object de”
Salient Object Detection in the Deep Learning Era: An In-Depth Survey (2019, methodology): “In computer graphics, SOD has been used tasks such as non-photo-realist rendering [15], [16], automatic image cropping [17], image retargeting [18], [19], video summarization [20], [21], etc .”