Cosegmentation of Image Pairs by Histogram Matching - Incorporating a Global Constraint into MRFs

By Carsten Rother, Tom Minka, Andrew Blake, Vladimir Kolmogorov. 2006

600 citations recorded by the source corpus.

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

SIFT Flow: Dense Correspondence across Scenes and Its Applications (2011, background): “5, we show how to infer the motion field from a single image, and how to animate a still image, both with the support of a large video database and scene alignment.”
Unsupervised object discovery and localization in the wild: Part-based matching with bottom-up region proposals (2015, background): “[38] who fuse Markov random fields with color histogram matching to segment objects common to two images.”
BiCoS: A Bi-level co-segmentation method for image classification (2011, background): “A number of approaches [16, 25, 31, 37] work with pixel RGB values to model and propagate the distributions of visual appearance.”
Co-Separating Sounds of Visual Objects (2019, background): “We call our idea “co-separation” as a loose analogy to image co-segmentation [38], whereby jointly segmenting two related images can be easier than segmenting them separately, since it allows disentangling a shared foreground object from differently cluttered backgrounds.”
GeoS: Geodesic Image Segmentation (2008, background): “The global soft constraint G cannot be written as a sum of unary and pairwise terms [ 23 ,24].,For example in [ 23 ] G = G(h1 ,h 2) is defined as a divergence between region histograms hi. General energy models of this kind have not been used much in the literature because of the lack of appropriate”
Discriminative clustering for image co-segmentation (2010, methodology): “In Section 4.1, we ﬁrst consider images with foreground objects which are identical or very similar in appearance and with few images to co-segment, a setting that was already used in [19] and extended in [12].”
Balloon Shapes: Reconstructing and Deforming Objects with Volume from Images (2013, methodology): “We could also rely on cosegmentation methods [20] to segment the same object in both images, or use a similar strategy to [9] alternating between estimating the shape and the segmentation.”
Image Co-segmentation via Consistent Functional Maps (2013, methodology): “Earlier work on joint segmentation mainly compared the visual features of image pairs, such as foreground color histogram [18], SIFT [14], saliency [5], and Gabor features [9].”
Pedestrian Attribute Recognition At Far Distance (2014, background): “It is worth noting that MRF inference for smoothing [9] is commonly applied in image segmentation [15], but this paper is the first work that explores this approach for pedestrian attribute inference.”
Unsupervised Multi-class Joint Image Segmentation (2014, neutral): “[14], who used color histogram matching to find common objects in a pair of images.”
Enriching Visual Knowledge Bases via Object Discovery and Segmentation (2014, background): “Our work is also related to co-segmentation, where the task is to simultaneously segment visually similar objects from multiple images at the same time [29, 28, 3, 11, 13, 36].”
Unsupervised Object Segmentation by Redrawing (2019, background): “Co-segmentation, introduced by Rother et al. 2006 [46], addresses the related problem of segmenting objects that are shared by multiple images by looking for similar data patterns in all those images.”
Salient object detection: A survey (2014, background): “ient object detection models only need a single input image. Co-saliency detection is closely related to the concept of image co-segmentation that aims to segment similar objects from multiple images [16,172]. As stated in [61], three major differences exist between co-saliency and cosegmentation. F”
ImageNet Auto-Annotation with Segmentation Propagation (2014, methodology): “Following [53], we now use this segmentation to update the appearance models, and then iteratively alternate between these two steps: finding the optimal segmentation x given the appearance models, and updating the appearance models given the segmentation.”
CRNet: Cross-Reference Networks for Few-Shot Segmentation (2020, methodology): “[15] propose to minimize an energy function of a histogram matching term with an MRF to enforce similar foreground statistics.”