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

Improving subspace learning for facial expression recognition using person dependent and geometrically enriched

Anastasios Maronidis1, Dimitris Bolis, Anastasios Tefas

  • 1Aristotle University of Thessaloniki, Department of Informatics, Box 451, 54124 Thessaloniki, Greece.

Neural Networks : the Official Journal of the International Neural Network Society
|August 9, 2011
PubMed
Summary

Related Concept Videos

Muscles for Facial Expressions01:14

Muscles for Facial Expressions

The craniofacial muscles are a collection of approximately 20 thin skeletal muscles situated beneath the skin of the face and scalp. These muscles, primarily responsible for the vast array of human facial expressions, originate from the bones or fibrous structures of the skull and extend outwards to connect with the skin. While most skeletal muscles in the body are enveloped in thick fascia, facial muscles generally have a more delicate fascial covering, with the buccinator muscle being a...

You might also read

Related Articles

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

Sort by
Same author

Transcutaneous vagus nerve stimulation modulates adaptive empathic learning under uncertainty.

Cortex; a journal devoted to the study of the nervous system and behavior·2026
Same author

Emotional egocentricity bias is modulated by implicit expectations of interpersonal emotional contingencies and perceptual noise.

Cognition·2026
Same author

Photonic Kolmogorov-Arnold networks based on self-phase modulation in nonlinear waveguides.

Optics letters·2026
Same author

LIX: Implicitly Infusing Spatial Geometric Prior Knowledge Into Visual Semantic Segmentation for Autonomous Driving.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2025
Same author

Comparison of ChatGPT-3.5 and GPT-4 as potential tools in artificial intelligence-assisted clinical practice in renal and liver transplantation.

World journal of transplantation·2025
Same author

Inducing Neural Collapse via Anticlasses and One-Cold Cross-Entropy Loss.

IEEE transactions on neural networks and learning systems·2025

This study examines how geometric transformations affect appearance-based subspace learning for facial expression recognition. Enriching training data with transformed images and using person-dependent training significantly improves recognition accuracy.

Area of Science:

  • Computer Vision
  • Machine Learning
  • Pattern Recognition

Background:

  • Appearance-based subspace learning methods are widely used for facial expression recognition.
  • These methods are known to be sensitive to geometric transformations like translation, scaling, and rotation.
  • Existing literature lacks systematic studies on the impact of these transformations on recognition accuracy.

Purpose of the Study:

  • To systematically evaluate the robustness of appearance-based subspace learning techniques against geometrical image transformations.
  • To investigate the correlation between image registration error and recognition accuracy in facial expression recognition.
  • To propose effective strategies for enhancing the robustness of these techniques.

Main Methods:

  • Tested several appearance-based subspace learning techniques on four facial expression databases.

Related Experiment Videos

  • Analyzed the relationship between recognition accuracy and image registration error.
  • Experimented with training set enrichment by including translated, scaled, and rotated images.
  • Compared person-dependent training with generic learning approaches.

    • Main Results:

    • A strong correlation was observed between recognition accuracy and image registration error.
    • Training set enrichment with geometrically transformed images was found to improve robustness.
    • Person-dependent training demonstrated significantly higher accuracy than generic learning for facial expression recognition.

    Conclusions:

    • Appearance-based subspace learning methods exhibit limited robustness to geometrical transformations.
    • Enriching training datasets with transformed images is a viable strategy to enhance robustness in facial expression recognition.
    • Person-dependent training offers superior performance compared to generic training for this task.