Humans recognize basic facial expressions effortlessly. Yet, despite a considerable amount of research, this task remains elusive for computer vision systems. Here, we compared the behavior of one of the best computer models of facial expression recognition (Z. Hammal, L. Couvreur, A. Caplier, & M. Rombaut, 2007) with the behavior of human observers during the M. Smith, G. Cottrell, F. Gosselin, and P. G. Schyns (2005) facial expression recognition task performed on stimuli randomly sampled using Gaussian apertures. The model--which we had to significantly modify in order to give the ability to deal with partially occluded stimuli--classifies the six basic facial expressions (Happiness, Fear, Sadness, Surprise, Anger, and Disgust) plus Neutral from static images based on the permanent facial feature deformations and the Transferable Belief Model (TBM). Three simulations demonstrated the suitability of the TBM-based model to deal with partially occluded facial parts and revealed the differences between the facial information used by humans and by the model. This opens promising perspectives for the future development of the model.
Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.1167/9.2.22 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Retinal Racemosal Hemangioma Associated with Intracranial Vascular Malformation (Wyburn -Mason Syndrome) - Case Report.
Retin Cases Brief Rep
December 2024
Vision and Ocular Health Research Group (VISOC) / Vision and Ocular Health Research Group (VISOC), Ophthalmology Service, Universidad del Valle, Hospital Universitario del Valle, Cali, Colombia.
Purpose: To describe a case of Wyburn-Mason syndrome in a young child with initially declined treatment leading to delayed intervention.
Methodology: Case report.
Results: We report a 6-year-old boy with mild periocular trauma following a fall, who was incidentally found to have a retinal arteriovenous malformation (AVM) associated with a brain AVM, indicative of Wyburn-Mason Syndrome.
On the influence of artificially distorted images in firearm detection performance using deep learning.
PeerJ Comput Sci
October 2024
Department of Computer Science and Statistics, Universidad Rey Juan Carlos, Mostoles, Madrid, Spain.
Detecting people carrying firearms in outdoor or indoor scenes usually identifies (or avoids) potentially dangerous situations. Nevertheless, the automatic detection of these weapons can be greatly affected by the scene conditions. Commonly, in real scenes these firearms can be seen from different perspectives.
View Article and Find Full Text PDFTranscatheter Pulmonary Flow Restrictors: Current Trends and Future Perspectives.
Catheter Cardiovasc Interv
January 2025
Pediatrics & Medicine, Weill Cornell Medicine Chief International Patients Services & Chair, Sidra Department Cardiovascular Diseases, University of Jordan, Amman, Jordan.
Article Synopsis
- Transcatheter Pulmonary Flow Restrictors (TPFRs) are innovative devices for managing pulmonary blood flow in congenital heart disease, yet there is a lack of detailed studies examining their variety and crucial design features.
- * The review consolidates current knowledge on TPFRs, highlights significant design considerations, identifies gaps in medical practice, and suggests future research avenues to improve these devices.
- * An analysis of 82 TPFR implants revealed challenges like complications with microvascular plugs, including issues with sizing, device migration, and thrombosis, while emphasizing the need for devices that can be safely removed and reduce risks like embolization and inflammation.
Improving facial expression recognition for autism with IDenseNet-RCAformer under occlusions.
Int J Dev Neurosci
February 2025
Department of Computer Science and Engineering Sethu Institute of Technology, Virudhunagar, Tamil Nadu, India.
Article Synopsis
- The text discusses autism spectrum disorder, focusing on the challenges faced in recognizing facial expressions due to occlusions like glasses or facial hair.
- It proposes a new system called the Improved DenseNet-based Residual Cross-Attention Transformer (IDenseNet-RCAformer) aimed at improving facial expression recognition for individuals with autism.
- The IDenseNet-RCAformer has shown impressive results, achieving an accuracy of 98.95% in identifying facial expressions despite partial occlusions, aided by advanced techniques for feature extraction and fusion.
Pulmonary Vein Thrombosis: An Unlikely Cause of Chest Pain.
Cureus
November 2024
Cardiology, Bruce Carter Miami VA (Veterans Affairs) Healthcare System, Florida, USA.
Pulmonary vein thrombosis (PVT) is a rare but potentially lethal source of arterial thromboembolism. We present a case of a 59-year-old female patient who presented with worsening shortness of breath and was found on a cardiac computed tomography angiography (CTA) with the right lower pulmonary vein partially occluded with a filling defect. This case report highlights the diagnostic challenges and management options associated with this rare entity.
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!