This paper presents a fast and efficient computational approach to higher order spectral graph matching. Exploiting the redundancy in a tensor representing the affinity between feature points, we approximate the affinity tensor with the linear combination of Kronecker products between bases and index tensors. The bases and index tensors are highly compressed representations of the approximated affinity tensor, requiring much smaller memory than in previous methods, which store the full affinity tensor. We compute the principal eigenvector of the approximated affinity tensor using the small bases and index tensors without explicitly storing the approximated tensor. To compensate for the loss of matching accuracy by the approximation, we also adopt and incorporate a marginalization scheme that maps a higher order tensor to matrix as well as a one-to-one mapping constraint into the eigenvector computation process. The experimental results show that the proposed method is faster and requires smaller memory than the existing methods with little or no loss of accuracy.
Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.1109/TPAMI.2013.157 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
The High-Affinity IL-2 Receptor Affects White Matter Damage after Cerebral Ischemia by Regulating CD8 + T Lymphocyte Differentiation.
J Neuroimmune Pharmacol
January 2025
Institute of Cerebrovascular Disease Research, Xuanwu Hospital of Capital Medical University, Beijing, 100053, China.
IL-2/IL-2R inhibition improved the prognosis of ischemic stroke by regulating T cells, while the respective contribution of T cells with high/medium/low-affinity IL-2 receptors remained unclear. Single-cell RNA sequencing data of ischemic brain tissue revealed that most of the high-affinity IL-2R would be expressed by CD8 + T cells, especially by a highly-proliferative subset. Interestingly, only the CD8 + T cells with high-affinity IL-2R infiltrated ischemic brain tissues, highly expressing 32 genes (including Cdc20, Cdca3/5, and Asns) and activating 7 signaling pathways (including the interferon-alpha response pathway, a key mediator in the proliferation, migration, and cytotoxicity of CD8 + T cells).
View Article and Find Full Text PDFA 4D tensor-enhanced multi-dimensional convolutional neural network for accurate prediction of protein-ligand binding affinity.
Mol Divers
December 2024
Laboratory of Molecular Design and Drug Discovery, School of Science, China Pharmaceutical University, 639 Longmian Avenue, Nanjing, 211198, China.
Protein-ligand interactions are the molecular basis of many important cellular activities, such as gene regulation, cell metabolism, and signal transduction. Protein-ligand binding affinity is a crucial metric of the strength of the interaction between the two, and accurate prediction of its binding affinity is essential for discovering drugs' new uses. So far, although many predictive models based on machine learning and deep learning have been reported, most of the models mainly focus on one-dimensional sequence and two-dimensional structural characteristics of proteins and ligands, but fail to deeply explore the detailed interaction information between proteins and ligand atoms in the binding pocket region of three-dimensional space.
View Article and Find Full Text PDFCSF levels of brain-derived proteins correlate with brain ventricular volume in cognitively healthy 70-year-olds.
Clin Proteomics
December 2024
Division of Affinity Proteomics, Department of Protein Science, KTH Royal Institute of Technology, SciLifeLab, Stockholm, Sweden.
Background: The effect of varying brain ventricular volume on the cerebrospinal fluid (CSF) proteome has been discussed as possible confounding factors in comparative protein level analyses. However, the relationship between CSF volume and protein levels remains largely unexplored. Moreover, the few existing studies provide conflicting findings, indicating the need for further research.
View Article and Find Full Text PDFPartition-level fusion induced multi-view Subspace Clustering with Tensorial Geman Rank.
Neural Netw
February 2025
School of Computer Science and Technology, Beijing Jiaotong University, Beijing, 100044, China; Engineering Research Center of Integration and Application of Digital Learning Technology, Ministry of Education, Beijing, China. Electronic address:
The tensor-based multi-view clustering approach captures the high-order correlation among different views by learning a low-rank representation tensor, which has achieved favorable performance in multi-view clustering. However, the tensor rank approximation functions used by the extant algorithms are not tight enough to the true rank of the tensor, leading to the undesired low-rank structure. Besides, the fusion strategy at the affinity matrix level is less robust to noise, resulting in sub-optimal clustering results.
View Article and Find Full Text PDFTensorial multiview low-rank high-order graph learning for context-enhanced domain adaptation.
Neural Netw
January 2025
Department of Computer Science, Durham University, United Kingdom.
Unsupervised Domain Adaptation (UDA) is a machine learning technique that facilitates knowledge transfer from a labeled source domain to an unlabeled target domain, addressing distributional discrepancies between these domains. Existing UDA methods often fail to effectively capture and utilize contextual relationships within the target domain. This research introduces a novel framework called Tensorial Multiview Low-Rank High-Order Graph Learning (MLRGL), which addresses these challenges by learning high-order graphs constrained by low-rank tensors to uncover contextual relations.
View Article and Find Full Text PDFWant AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!