A probabilistic spectral framework for grouping and segmentation

Abstract

This paper presents an iterative spectral framework for pairwise clustering and perceptual grouping. Our model is expressed in terms of two sets of parameters. Firstly, there are cluster memberships which represent the a nity of objects to clusters. Secondly, there is a matrix of link weights for pairs of tokens. We adopt a model in which these two sets of variables are governed by a Bernoulli model. We show how the likelihood function resulting from this model may be maximised with respect to both the elements of link-weight matrix and the cluster membership variables. We establish the link between the maximisation of the log-likelihood function and the eigenvectors of the link-weight matrix. This leads us to an algorithm in which we iteratively update the link-weight matrix by repeatedly reÿning its modal structure. Each iteration of the algorithm is a three-step process. First, we compute a link-weight matrix for each cluster by taking the outer-product of the vectors of current cluster-membership indicators for that cluster. Second, we extract the leading eigenvector from each modal link-weight matrix. Third, we compute a revised link weight matrix by taking the sum of the outer products of the leading eigenvectors of the modal link-weight matrices.

Motion features to enhance scene segmentation in active visual attention

“…(Robles-Kelly and Hancock, 2004), and also comparing the results with the Hamburg Taxi sequence, we observe that again our algorithm loses one car due to the imposed shape size criteria (leading to the union of the tree with the car). Following the terminology of Robles-Kelly and Hancock, our algorithm does not detect one of the three clusters.…”

MoE-SPNet: A mixture-of-experts scene parsing network

“…In the past decade, the successful scene parsing methods rely on handcrafted local features like colour histogram and textons [6,29,30,31,32,33], and shallow classifiers such as Boosting [6,34], Random Forests [35,36], Support Vector Machines [37]. Due to the limited discriminative power of local features, a lot of efforts have been put into developing probabilistic graphical models such as CRFs to enforce spatial consistency and incorporate rich contextual information [38,7,39,40]. Recently, deep learning methods typified by DCNNs have achieved state-of-the-art performance on various computer vision tasks, such as image classification and multi-class object detection.…”

A Thermodynamics Approach to Graph Similarity

“…We commence the indexing process applying the pairwise clustering clustering algorithm of Robles-Kelly and Hancock [9] to the similarity data for the complete set of graphs in the database. The clustering algorithm requires distances to be represented by a matrix of pairwise affinity weights.…”