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

Combining reconstructive and discriminative subspace methods for robust classification and regression by subsampling.

Sanja Fidler1, Danijel Skocaj, Ales Leonardis

  • 1Faculty of Computer and Information Science, University of Ljubljana, Slovenia. sanja.fidler@fri.uni-lj.si

IEEE Transactions on Pattern Analysis and Machine Intelligence
|March 11, 2006
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

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

Sort by
Same author

Patherea: Cell detection and classification for the 2020s.

Medical image analysis·2025
Same author

Transformer-Based Models for Predicting Molecular Structures from Infrared Spectra Using Patch-Based Self-Attention.

The journal of physical chemistry. A·2025
Same author

Structure and Intensity Unbiased Translation for 2D Medical Image Segmentation.

IEEE transactions on pattern analysis and machine intelligence·2024
Same author

Weakly-Supervised RGBD Video Object Segmentation.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2024
Same author

FlexHDR: Modeling Alignment and Exposure Uncertainties for Flexible HDR Imaging.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2022
Same author

Deep neural networks and image classification in biological vision.

Vision research·2022
Same journal

HardFlow: Hard-Constrained Sampling for Flow-Matching Models Via Trajectory Optimization.

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

Industrial Brain: Self-Evolving Neuro-Symbolic Autonomy with Causal Resilience for Cyber-Physical Systems.

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

Adaptive Hardness-Driven Dictionary Distillation for Incomplete Streaming View Clustering.

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

Mixture of Global and Local Experts with Diffusion Transformer for Controllable Face Generation.

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

Task-KV: Task-aware KV Cache Optimization via Semantic Differentiation of Attention Heads.

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

Achieving Text-based Person Retrieval with Any Granularity.

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

This study introduces a novel framework combining data reconstruction and discrimination for robust image classification. The approach effectively handles missing data, outliers, and occlusions, outperforming standard methods.

Area of Science:

  • Computer Vision
  • Machine Learning
  • Data Science

Background:

  • Linear subspace methods like PCA excel at data reconstruction, handling missing pixels, outliers, and occlusions.
  • Discriminative methods like LDA are effective for classification but sensitive to corrupted data.

Purpose of the Study:

  • To develop a theoretical framework combining the strengths of reconstructive and discriminative methods.
  • To enable robust classification even with corrupted or incomplete image data.

Main Methods:

  • A novel approach integrating discriminative power with reconstructive properties.
  • Utilizing subsets of pixels for efficient outlier detection and rejection.
  • Extending the framework to regression tasks using methods like CCA.

Related Experiment Videos

Main Results:

  • The proposed approach achieves robust classification with a high breakdown point.
  • Demonstrated superior performance on computer vision tasks with missing pixels, occlusions, and outliers.
  • Outperformed standard discriminative methods in challenging data scenarios.

Conclusions:

  • The combined framework offers a powerful solution for robust image analysis and classification.
  • This approach enhances the reliability of machine learning models in real-world, imperfect data conditions.