Export Citations
Save this search
Please login to be able to save your searches and receive alerts for new content matching your search criteria.
- research-articleOctober 2024
Semantic-embedded similarity prototype for scene recognition
AbstractDue to the high inter-class similarity caused by the complex composition and the co-existing objects across scenes, numerous studies have explored object semantic knowledge within scenes to improve scene recognition. However, a resulting ...
Highlights- We propose a semantic-embedded similarity prototype to augment the network’s training process by providing prior knowledge. The trained network achieves superior recognition performance in real applications without the need for intensive ...
- research-articleOctober 2024
Large-scale multi-view clustering via matrix factorization of consensus graph
AbstractRecently, anchors-based multi-view clustering methods have been widely concerned for they can not only significantly reduce the time complexity but also have good interpretability. However, the time consumption of optimization and spectral ...
Graphical abstractDisplay Omitted
Highlights- The transition matrix minimizes clustering time growth with more anchors.
- Optimized low-dimension embedding matrix exploits consensus without SVD operations.
- Proven algorithm improves clustering performance, comparable in time ...
- research-articleOctober 2024
View-unaligned clustering with graph regularization
AbstractIn current multi-view clustering modeling scenarios, the cross-view correspondence of the data is generally presumed in advance. However, this assumption is inevitably violated in practical applications as each view is independently processed ...
Highlights- We propose a novel clustering solution for addressing the view-unaligned problem.
- The proposed alignment scheme eliminates the reliance on the partially aligned data.
- Latent embedding learning, alignment and partition are ...
- research-articleOctober 2024
Robust Self-expression Learning with Adaptive Noise Perception
AbstractSelf-expression learning methods often obtain a coefficient matrix to measure the similarity between pairs of samples. However, directly using the raw data to represent each sample under the self-expression framework may not be ideal, as noise ...
Highlights- A novel concept of noise perception is presented.
- Noise points and the noisy part of each sample are detected adaptively.
- Only clean samples are used to realize self-expression learning.
- The properties of dynamic learning of ...
- research-articleOctober 2024
Dual space-based fuzzy graphs and orthogonal basis clustering for unsupervised feature selection
AbstractUnsupervised feature selection (UFS) takes an important position because gaining the class labels is laborious or even impossible. In the domain of UFS, clustering is a major means to exploit label information. The existing methods either could ...
Highlights- Orthogonal basis clustering and the fuzzy graph are integrated.
- The dual-graph retains more comprehensive local manifold structures.
- l 2 , 0-norm is applied to ensure feature selection’s sparsity and accuracy.
-
- research-articleOctober 2024
G2-SCANN: Gaussian-kernel graph-based SLD clustering algorithm with natural neighbourhood
Highlights- The shortest path length (SPL) in complex network or graph-based geodesic distance is used to give a locally backbone-structured description of graph vertex similarity. Accordingly, SPL-weighted local degree (SLD) is defined as vertex ...
For most clustering methods, not only the number of clusters must be set in advance, but also various hyperparameters such as initial centroids, number of nearest neighbours, the minimum number of points, neighbourhood radius, and cutoff distance ...
- research-articleOctober 2024
A lie group semi-supervised FCM clustering method for image segmentation
AbstractAs an unsupervised clustering method with low overhead, Fuzzy C-means (FCM) clustering has been widely used in a variety of image segmentation tasks. However, existing FCM clustering methods are sensitive to image noises and are either suffer ...
Highlights- Lie group features containing weighted average filtering information are constructed to summarize local contextual information of pixels and global relationships between features.
- A small number of labeled pixels are transformed to ...
Display Omitted
- research-articleOctober 2024
Multi-view reduced dimensionality K-means clustering with σ-norm and Schatten p-norm
AbstractRecently, multi-view high dimensional data obtained from diverse domains or various feature extractors has drawn great attention due to its reflection of different properties or distributions. In this paper, we propose a novel unsupervised multi-...
Highlights- In order to avoid the influence of dimensional curse and redundant features in the original space, we use dimension reduction technology to process high-dimensional multi-view data.
- We use the σ-norm as an adaptive loss minimization, ...
- research-articleOctober 2024
EDMD: An Entropy based Dissimilarity measure to cluster Mixed-categorical Data
AbstractThe effectiveness of clustering techniques is significantly influenced by proximity measures irrespective of type of data and categorical data is no exception. Most of the existing proximity measures for categorical data assume that all ...
Highlights- A new entropy-based proximity measure is proposed for clustering mixed categorical data.
- Nominal and ordinal attributes treated separately using Boltzmann’s definition of entropy.
- The significance of attributes is taken care of ...
- research-articleOctober 2024
Contrastive cross-modal clustering with twin network
AbstractCross-modal clustering (CMC) methods explore the correlation information between multiple modalities to improve clustering performance. However, the obvious differences between heterogeneous modalities make it difficult to obtain the correlation ...
Highlights- In this paper, a novel 3CTnet method is proposed for cross-modal clustering.
- We contrast the differences of multiple modalities to mine correlation information.
- A correlation propagate module is designed to propagate the ...
- research-articleOctober 2024
A novel K-means and K-medoids algorithms for clustering non-spherical-shape clusters non-sensitive to outliers
AbstractDetermination of the optimal number of clusters, the random selection of the initial centers, the non-detection of non-spherical clusters, and the negative impact of outliers are the main challenges of the K-means algorithm. In this paper, to ...
Highlights- An Intelligent center selection method is proposed for faster convergence.
- A method is proposed to identify clusters with non-spherical and asymmetrical shapes.
- Two selection methods are proposed to determine the other initial ...
- research-articleOctober 2024
Efficient and robust clustering based on backbone identification
AbstractClustering is the process of grouping similar data objects into different subsets based on their similarities. Inspired by the concept of the popularity of individuals in a community, we rate the popularity of each sample which reflects the ...
Highlights- A method is proposed for assigning popularity to samples.
- A scale-independent backbone identification method is proposed.
- A popularity-based clustering algorithm is proposed.
- A new popularity-based proximity measure is ...
- research-articleOctober 2024
Linear Centroid Encoder for Supervised Principal Component Analysis
AbstractWe propose a new supervised dimensionality reduction technique called Supervised Linear Centroid-Encoder (SLCE), a linear counterpart of the nonlinear Centroid-Encoder (CE) (Ghosh and Kirby, 2022). SLCE works by mapping the samples of a class to ...
Highlights- New technique for linear dimensionality reduction: Supervised Linear Centroid-Encoder.
- The method uses class centroids instead of labels to impose supervision in learning.
- Proposed a closed form solution using eigendecomposition.
- research-articleOctober 2024
Graph domain adaptation with localized graph signal representations
AbstractIn this paper we propose a domain adaptation algorithm designed for graph domains. Given a source graph with many labeled nodes and a target graph with few or no labeled nodes, we aim to estimate the target labels by making use of the similarity ...
Highlights- A new method is proposed for domain adaptation on graphs.
- The algorithm is based on considering label functions as graph signals.
- The label function is represented in terms of spectral graph wavelets.
- The wavelet coefficients ...
- research-articleMarch 2024
Learning a target-dependent classifier for cross-domain semantic segmentation: Fine-tuning versus meta-learning
Highlights- The introduction of a target-dependent classifier for cross-domain semantic segmentation able to better fit with the target domain features even under imperfect domain alignment.
- An innovative linkage of meta-learning with domain ...
Recently proposed domain adaptation arts have dominated the field of cross-domain semantic segmentation by operating domain manifolds alignment and learning an optimal joint hypothesis (joint-domain classifier) for both source and target domains. ...
- research-articleMarch 2024
Geometric-inspired graph-based Incomplete Multi-view Clustering
AbstractMulti-view clustering methods group data into different clusters by discovering the consensus in heterogeneous sources, which however becomes difficult when partial views of real-world data are missing. Consequently, reducing the impact of ...
Highlights- We conduct geometric analyses to mitigate missing views in weight aggregation.
- Our analyses aid graph aggregation for incomplete multi-view clustering.
- Experiments show our method’s effectiveness and flexibility in weight ...
- research-articleFebruary 2024
Deep image clustering with contrastive learning and multi-scale graph convolutional networks
AbstractDeep clustering has shown its promising capability in joint representation learning and clustering via deep neural networks. Despite the significant progress, the existing deep clustering works mostly utilize some distribution-based clustering ...
Highlights- This paper for the first time enables multi-scale structure learning for image clustering via GCNs.
- Multi-scale structure learning and two levels of contrastive learning are jointly enforced.
- A novel deep image clustering approach ...
- research-articleFebruary 2024
GNaN: A natural neighbor search algorithm based on universal gravitation
AbstractThe natural neighbor (NaN) method and its search algorithm (NaN-Searching) are widely used in many fields, including pattern recognition and image processing. NaN-Searching fundamentally overcomes the problem of the conventional nearest neighbor ...
Highlights- The problem of parameter selection is solved.
- It can better reflect the overall characteristics of the dataset.
- Better adapt to datasets with complex manifold structures.
- research-articleFebruary 2024
Coordinate Descent Optimized Trace Difference Model for Joint Clustering and Feature Extraction
AbstractJoint clustering and dimensionality reduction methods are a promising solution to clustering due to its scalability to high-dimensional data. Some methods leverage trace ratio criterion and attain clusters by borrowing the K-means algorithm. ...
Highlights- A joint clustering and feature extraction model named CDOTD is proposed.
- It ensures both favorable clustering performance and high running efficiency.
- It improves the ability to address many cluster data clustering cases.
- ...
- research-articleFebruary 2024
Beyond k-Means++: Towards better cluster exploration with geometrical information
AbstractAlthough k-means and its variants are known for their remarkable efficiency, they suffer from a strong dependence on the prior knowledge of K and the assumption of a circle-like pattern, which can result in the algorithms dividing the input space ...
Highlights- A novel framework of iterative division and aggregation (IDA) over k-means++ works with any K.
- A reasonability checking strategy (RCS) makes beyond k-means++ support arbitrary cluster shapes.
- An edge shrinkage strategy (ESS) allows ...