Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Deep learning with hierarchical convolutional factor analysis.

Bo Chen1, Gungor Polatkan, Guillermo Sapiro

  • 1Electrical and Computer Engineering Department, Duke University, 130 Hudson Hall, Durham, NC 27708-0291, USA.

IEEE Transactions on Pattern Analysis and Machine Intelligence
|June 22, 2013
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

Factorial Design02:01

Factorial Design

Factorial Analysis is an experimental design that applies Analysis of Variance (ANOVA) statistical procedures to examine a change in a dependent variable due to more than one independent variable, also known as factors. Changes in worker productivity can be reasoned, for example, to be influenced by salary and other conditions, such as skill level. One way to test this hypothesis is by categorizing salary into three levels (low, moderate, and high) and skills sets into two levels (entry level...

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Environment and reproductive health in China: challenges and opportunities.

Environmental health perspectives·2012
Same author

Posttransplant mortality risk assessment for adult-to-adult right-lobe living donor liver recipients with benign end-stage liver disease.

Scandinavian journal of gastroenterology·2012
Same author

Sodium nitrite protects against kidney injury induced by brain death and improves post-transplant function.

Kidney international·2012
Same author

OIC-A006-loaded true bone ceramic heals rabbit critical-sized segmental radial defect.

Die Pharmazie·2012
Same author

Liquid chromatography-mass spectrometric multiple reaction monitoring-based strategies for expanding targeted profiling towards quantitative metabolomics.

Current drug metabolism·2012
Same author

Structural and functional characterization of mature forms of metalloprotease E495 from Arctic sea-ice bacterium Pseudoalteromonas sp. SM495.

PloS one·2012
Same journal

Relation DETR+: Exploring Explicit Position Relation Prior for Dense Prediction.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

RBF++: Quantifying and Optimizing Reasoning Boundaries across Measurable and Unmeasurable Capabilities for Chain-of-Thought Reasoning.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

CAFE: Cross-View Adaptive Fusion and Cluster Center Enhancement for Robust Multi-View Clustering.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

DIVER: Reinforced Diffusion Breaks Imitation Bottlenecks in End-to-End Autonomous Driving.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

Ethics-Aware Safe Reinforcement Learning for Rare-Event Risk Control in Interactive Urban Driving.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

Learning Shape Anchors for Holistic Indoor Scene Understanding.

IEEE transactions on pattern analysis and machine intelligence·2026
See all related articles

This study introduces unsupervised deep models for image processing using hierarchical convolutional factor analysis. The novel Bayesian approach infers model parameters and dictionary elements from data, enabling efficient analysis of large-scale image datasets.

Area of Science:

  • Computer Vision
  • Machine Learning
  • Statistical Modeling

Background:

  • Deep unsupervised models are crucial for image processing.
  • Hierarchical convolutional factor analysis offers a structured approach.
  • Bayesian methods provide a robust framework for parameter estimation.

Purpose of the Study:

  • To develop and evaluate unsupervised deep models for imagery.
  • To implement a hierarchical convolutional factor-analysis construction.
  • To enable efficient processing of large-scale and streaming image data.

Main Methods:

  • Utilizing a hierarchical convolutional factor-analysis construction with sparse loadings and scores.
  • Employing Bayesian inference with Gibbs sampling and variational Bayesian (VB) analysis.

Related Experiment Videos

  • Developing an online VB version for large-scale and streaming data.
  • Inferring dictionary elements using a beta-Bernoulli Indian buffet process.

    • Main Results:

    • Demonstrated effectiveness of the proposed model in image-processing applications.
    • Showcased the ability to infer layer-dependent parameters and dictionary sizes.
    • Presented comparisons with existing related models in the literature.

    Conclusions:

    • The developed unsupervised deep models are effective for image processing.
    • The Bayesian framework and online VB offer scalable solutions for complex image data.
    • The model successfully infers structural properties from the data.