Learning and inferring image segmentations using the GBP typical cut algorithm

Shental,; Zomet,; Hertz,; Weiss,

doi:10.1109/iccv.2003.1238633

Cited by 30 publications

(24 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There are many recent works on multi-class image segmentation that employ some kind of contextual information (e.g., [4,9,10,12,19,22,23,24,29,31]). The simplest type of contextual information is in the form of a continuity preference for nearby pixels.…”

Section: Related Workmentioning

confidence: 99%

“…Max-product loopy belief propagation is one such method. Here messages are passed between superpixels to iteratively update each superpixel's belief over its class label distribution (e.g., [23]). The method is simple to implement and its use is supported by the findings of Szeliski et al [27] who show that, even in the context of binary image segmentation, using loopy belief propagation incurs only 0.1% degradation in performance when compared to the exact solution.…”

Section: Probabilistic Segmentationmentioning

confidence: 99%

See 1 more Smart Citation

Multi-Class Segmentation with Relative Location Prior

et al. 2008

View full text Add to dashboard Cite

Multi-class image segmentation has made significant advances in recent years through the combination of local and global features. One important type of global feature is that of inter-class spatial relationships. For example, identifying "tree" pixels indicates that pixels above and to the sides are more likely to be "sky" whereas pixels below are more likely to be "grass." Incorporating such global information across the entire image and between all classes is a computational challenge as it is image-dependent, and hence, cannot be precomputed.In this work we propose a method for capturing global information from inter-class spatial relationships and encoding it as a local feature. We employ a two-stage classification process to label all image pixels. First, we generate predictions which are used to compute a local relative location feature from learned relative location maps. In the second stage, we combine this with appearance-based features to provide a final segmentation. We compare our results to recent published results on several multiclass image segmentation databases and show that the incorporation of relative location information allows us to significantly outperform the current state-of-the-art.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Probabilistic Segmentationmentioning

confidence: 99%

Multi-Class Segmentation with Relative Location Prior

et al. 2008

View full text Add to dashboard Cite

show abstract

“…The main research directions for this include mode-seeking [1], [2], deterministic annealing [3], stochastic clustering [4], [5], mixture model [6] [7], rate distortion [8], graph-based model [9], [10], [11], contourbased model [12], [13], and other variational methods [14], [15]. In most researches, the image segmentation problem is described as assigning a label to every pixel in a specific globalization framework.…”

Section: Introductionmentioning

confidence: 99%

Learning full pairwise affinities for spectral segmentation

Kim

Lee

2010

2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition

107

View full text Add to dashboard Cite

Abstract-Segmenting a single image into multiple coherent groups remains a challenging task in the field of computer vision. Particularly, spectral segmentation which uses the global information embedded in the spectrum of a given image's affinity matrix is a major trend in image segmentation. This paper focuses on the problem of efficiently learning a full range of pairwise affinities gained by integrating local grouping cues for spectral segmentation. We first construct a sparse multilayer graph whose nodes are both the pixels and the oversegmented regions obtained by an unsupervised segmentation algorithm. By applying the semi-supervised learning strategy to this graph, the intra and interlayer affinities between all pairs of nodes can be estimated without iteration. These pairwise affinities are then applied into the spectral segmentation algorithms. In this paper, two types of spectral segmentation algorithms are introduced: K-way segmentation and hierarchical segmentation. Our algorithms provide high-quality segmentations which preserve object details by directly incorporating the full-range connections. Moreover, since our full affinity matrix is defined by the inverse of a sparse matrix, its eigendecomposition can be efficiently computed. The experimental results on the BSDS and MSRC image databases demonstrate the superiority of our segmentation algorithms in terms of relevance and accuracy compared with existing popular methods.

show abstract

“…There are several segmentation methods that depend on the agreement between multiple segmentations of the image that differ from one another in some way (see for example Shental et al 2003, andCho andMeer 1997). The multiplicity of segmentations could be produced by one of the fast segmentation algorithms studied here by giving it the same amount of time taken by one of the slow segmentation algorithm, and allowing it to run with different parameter settings.…”

Section: Resultsmentioning

confidence: 99%

“…A final step of region merging to remove trivial cuts produces a segmentation of the image. Current algorithms related to SE-MinCut minimum-cut include Veksler (2000), Blake et al (2004), , Ishikawa and Geiger (1998), Wang and Siskind (2001), Gdalyahu et al (2001), Wang and Siskind (2003), Shental et al (2003), and of course, the normalized cuts algorithm described next. Shi and Malik (2000) propose that the optimal segmentation of the image corresponds to the partition of the graph that minimizes the normalized cut measure…”

Section: Spectral Embedding Mincutmentioning

confidence: 99%

Benchmarking Image Segmentation Algorithms

2009

View full text Add to dashboard Cite

We present a thorough quantitative evaluation of four image segmentation algorithms on images from the Berkeley Segmentation Database. The algorithms are evaluated using an efficient algorithm for computing precision and recall with regard to human ground-truth boundaries. We test each segmentation method over a representative set of input parameters, and present tuning curves that fully characterize algorithm performance over the complete image database. We complement the evaluation on the BSD with segmentation results on synthetic images. The results reported here provide a useful benchmark for current and future research efforts in image segmentation.

show abstract

Learning and inferring image segmentations using the GBP typical cut algorithm

Abstract: Abstract

Cited by 30 publications

References 11 publications

Multi-Class Segmentation with Relative Location Prior

Multi-Class Segmentation with Relative Location Prior

Learning full pairwise affinities for spectral segmentation

Benchmarking Image Segmentation Algorithms

Contact Info

Product

Resources

About