Learning Segmentation by Random Walks

By Marina Meila, Jianbo Shi. Neural Information Processing Systems, 2000

476 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/af865931f873f1c4686dcd2d6b297705d9709c93

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

Alternative Objective Functions for Deep Clustering (2018, background): “Stochastic normalization: Another normalization comes from viewing the afﬁnities as a stochastic matrix, leading to the random-walk interpretation of spectral clustering [11].”
Spectral Clustering for Speech Separation (2009, methodology): “While K-means is often used heuristically as a post-processor for spectral clustering (Ng et al., 2002, Meila and Shi, 2002), our approach provides a mathematical foundation for the use of K-means.”
Learning affinity functions for image segmentation: combining patch-based and gradient-based approaches (2003, background): “It is a natural proposal [22] to learn optimal pairwise affinities from training data.”
Maximin affinity learning of image segmentation (2009, background): “There have been attempts to train affinity classifiers to produce good segmentations when partitioned by normalized cuts [17, 2].”
A survey of kernel and spectral methods for clustering (2008, methodology): “[20] [21] [22] [23].”
Three Systems of Insular Functional Connectivity Identified with Cluster Analysis (2010, methodology): “We chose to implement the basic k-means algorithm rather than more sophisticated graph-theoretic or spectral techniques (e.g., Meilă and Shi 2000; Shi and Malik 2000; Ng et al. 2002) because the latter are intended for sparse graphs.”
Learning Spectral Clustering (2003, methodology): “We study the cost function J(W, e), its approximation based on the power method presented in Section 3, and two existing app ro ches, one based on a Markov chain interpretation of spectral clustering [15] an d one based on the alignment [16] of DWD andΠ0.”
Self-Tuning Spectral Clustering (2004, background): “A second generation provided a level of theoretical analysis, and suggested improved algorithms (e.g., [6] [10] [5] [4] [3]).”
From region similarity to category discovery (2011, methodology): “We use spectral clustering [15] with affinities defined by a radial basis function (RBF) kernel on the learned distances:”
Towards a theoretical foundation for Laplacian-based manifold methods (2005, background): “1 Prior Work Many manifold and graph-motivated learning methods have been recently proposed, including [22, 27, 3, 12] for visualization and data representation, [30, 29, 9, 23, 4, 2, 26] for partially supervised classification and [25, 28, 24, 18, 14] among others for spectral clustering.”
A Random Walks View of Spectral Segmentation (2001, methodology): “We address this issues and propose a method for learning the optimal combination in [7].”
A Unified Framework for Model-based Clustering (2003, methodology): “Despite this disadvantage, discriminative methods such as graph partitioning and spectral clustering algorithms (Karypis et al., 1999; Dhillon, 2001; Meila and Shi, 2001; Ng et al., 2002; Strehl and Ghosh, 2002) have gained recent popularity due to their ability to produce desirable clustering resul”
Handbook of Cluster Analysis (2015, methodology): “The algorithm we recommend is based on [29, 30] and [34].”
Parallel Spectral Clustering in Distributed Systems (2011, methodology): “Recently, spectral clustering methods (e.g., [35], [31], [13]) which exploit pairwise similarities of data instances have been shown to be more effective than traditional methods such as k-means, which only considers the similarity values from instances to k centers.”
On Spectral Clustering: Analysis and an algorithm (2001, methodology): “We also note that this is only one of a large family of possible algorithms, and later discuss some related methods (e.g., [6]).”