Recovering hidden components in multimodal data with composite diffusion operators
From MaRDI portal
Abstract: Finding appropriate low dimensional representations of high-dimensional multi-modal data can be challenging, since each modality embodies unique deformations and interferences. In this paper, we address the problem using manifold learning, where the data from each modality is assumed to lie on some manifold. In this context, the goal is to characterize the relations between the different modalities by studying their underlying manifolds. We propose two new diffusion operators that allow to isolate, enhance and attenuate the hidden components of multi-modal data in a data-driven manner. Based on these new operators, efficient low-dimensional representations can be constructed for such data, which characterize the common structures and the differences between the manifolds underlying the different modalities. The capabilities of the proposed operators are demonstrated on 3D shapes and on a fetal heart rate monitoring application.
Recommendations
Cites work
- Bi-stochastic kernels via asymmetric affinity functions
- ConceFT: concentration of frequency and time via a multitapered synchrosqueezed transform
- Concerning nonnegative matrices and doubly stochastic matrices
- Diffusion maps
- Diffusion maps for changing data
- Diffusion maps, spectral clustering and reaction coordinates of dynamical systems
- Dynamic isoperimetry and the geometry of Lagrangian coherent structures
- scientific article; zbMATH DE number 1235882 (Why is no real title available?)
- Joint Manifolds for Data Fusion
- Laplacian Eigenmaps for Dimensionality Reduction and Data Representation
- Latent common manifold learning with alternating diffusion: analysis and applications
- Learning the geometry of common latent variables using alternating-diffusion
- Magnetic eigenmaps for the visualization of directed networks
- Map-based exploration of intrinsic shape differences and variability
- RELATIONS BETWEEN TWO SETS OF VARIATES
- Spectral convergence of the connection Laplacian from random samples
- The earth mover's distance as a metric for image retrieval
- The Wasserstein distance and approximation theorems
- Time coupled diffusion maps
- Wave-shape function analysis. When cepstrum meets time-frequency analysis
Cited in
(10)- Learning the geometry of common latent variables using alternating-diffusion
- Latent common manifold learning with alternating diffusion: analysis and applications
- Diffusion operators for multimodal data analysis
- Explore Intrinsic Geometry of Sleep Dynamics and Predict Sleep Stage by Unsupervised Learning Techniques
- scientific article; zbMATH DE number 2045463 (Why is no real title available?)
- Diffusion state distances: multitemporal analysis, fast algorithms, and applications to biological networks
- A continuum limit for the PageRank algorithm
- Spectral Discovery of Jointly Smooth Features for Multimodal Data
- Spatiotemporal analysis using Riemannian composition of diffusion operators
- Landmark alternating diffusion
This page was built for publication: Recovering hidden components in multimodal data with composite diffusion operators
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5025791)
