Image Segmentation with Ratio Cut

By Song Wang, Jeffrey Mark Siskind. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2003

335 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/e388fbe0f992cdf725b2950fa6cb36bfe9065305

How researchers see this paper

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

Citations in the corpus

Salient object detection: A survey (2014, methodology): “ed to form a weighted graph. This graph is partitioned into components by minimizing a cost function of the inter and/or intra components, e.g., Minimum Cut [167], Normalized Cut [141], and Ratio Cut [168]. A representative graph-based image segmentation algorithm is [91], which adopts a local optim”
Object segmentation using graph cuts based active contours (2003, background): “The normalized cut approach [19] is aimed at reducing this bias by introducing a cost function called disassociation, but it also introduces another bias towards similar weight partition [20].”
RIGOR: Reusing Inference in Graph Cuts for Generating Object Regions (2014, background): “However, the original min-cut approaches either tend to find foreground regions with a very small size [35] or attempt to solve an NP-hard problem [30] [33].”
Secrets of GrabCut and Kernel K-Means (2015, unknown): “(17)”
Color Image Segmentation Based on Mean Shift and Normalized Cuts (2007, methodology): “[8] S. Wang and J. M. Siskind, “Image segmentation with ratio cut,” IEEE Trans.”
Constrained parametric min-cuts for automatic object segmentation (2010, methodology): “Graph partition properties (8 features) include the cut (sum of affinity along the segment boundary) [35], the ratio cut (sum of affinity along the boundary divided by their number) [33], the normalized cut (ratio of cut and affinity inside foreground, plus ratio of cut and affinity in background) [”
A Formal Study of Shot Boundary Detection (2007, methodology): “Therefore, several other objectives such as ratio cut [53], normalized cut [54] and min-max cut [55] are proposed successively.”
Isoperimetric graph partitioning for image segmentation (2006, background): “There is a deep connection between electric circuits and random walks on graphs [19], [20], which suggests the analysis of this algorithm in terms of a random walk on a graph.”
Edge and line oriented contour detection: State of the art (2011, methodology): “Specifically, the most important techniques are based on graph theory and cuts [13,14], iterative mean shift [15], statistical methods [16,17], watershed [18,19], and multichannel analysis [20,21].”
A survey of graph theoretical approaches to image segmentation (2013, methodology): “Ratio cut [52] Rcut A,B ð Þ 1⁄4 cut1 A,B ð Þ”
CPMC: Automatic Object Segmentation Using Constrained Parametric Min-Cuts (2012, methodology): “(sum of affinities along the segment boundary) [67], the ratio cut (sum of affinity along the boundary divided by their number) [62], the normalized cut (ratio of cut and affinity inside foreground, plus ratio of cut and affinity on back-”
Optimal Surface Segmentation in Volumetric Images-A Graph-Theoretic Approach (2006, unknown): “Typical graph algorithms that were exploited for image segmentation include minimum spanning trees [16], [17], [18], [19], [20], shortest paths [7], [12], [21], [22], [23], [24], [25], [26], [27], and graph-cuts [28], [29], [30], [31], [32], [33], [34], [35], [36], [37].”