AI Article Synopsis
Article Abstract
A growing body of research suggests that movement aids facial expression recognition. However, less is known about the conditions under which the dynamic advantage occurs. The aim of this research was to test emotion recognition in static and dynamic facial expressions, thereby exploring the role of three featural parameters (prototypicality, ambiguity, and complexity) in human and machine analysis. In two studies, facial expression videos and corresponding images depicting the peak of the target and non-target emotion were presented to human observers and the machine classifier (FACET). Results revealed higher recognition rates for dynamic stimuli compared to non-target images. Such benefit disappeared in the context of target-emotion images which were similarly well (or even better) recognised than videos, and more prototypical, less ambiguous, and more complex in appearance than non-target images. While prototypicality and ambiguity exerted more predictive power in machine performance, complexity was more indicative of human emotion recognition. Interestingly, recognition performance by the machine was found to be superior to humans for both target and non-target images. Together, the findings point towards a compensatory role of dynamic information, particularly when static-based stimuli lack relevant features of the target emotion. Implications for research using automatic facial expression analysis (AFEA) are discussed.
Download full-text PDF |
Source |
|---|---|
| http://www.ncbi.nlm.nih.gov/pmc/articles/PMC10546417 | PMC |
| http://dx.doi.org/10.3389/fpsyg.2023.1221081 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Prototypical classifier with distribution consistency regularization for generalized category discovery: A strong baseline.
Neural Netw
February 2025
School of Computer Science and School of Artificial Intelligence, Optics and Electronics (iOPEN), Northwestern Polytechnical University, Beilin, Xi'an, 710072, Shaanxi, China.
Generalized Category Discovery (GCD) addresses a more realistic and challenging setting in semi-supervised visual recognition, where unlabeled data contains samples from both known and novel categories. Recently, prototypical classifier has shown prominent performance on this issue, with the Softmax-based Cross-Entropy loss (SCE) commonly employed to optimize the distance between a sample and prototypes. However, the inherent non-bijectiveness of SCE prevents it from resolving intraclass relations among samples, resulting in semantic ambiguity.
View Article and Find Full Text PDFA Novel Web App for Dietary Weight Management: Development, Implementation, and Usability Study.
JMIR Form Res
November 2024
Division of Nutritional Sciences, University of Illinois at Urbana Champaign, Urbana, IL, United States.
Background: Online weight loss programs have ambiguous efficacy. There is a growing body of evidence that weight loss programs when combined with apps have better outcomes; however, many apps lack an evidence-based approach to dietary changes for weight loss and do not rely on a theoretical framework for behavior change.
Objective: This study aimed to describe the development and the preliminary usability and acceptability testing of a web app that uses behavior change techniques (BCTs) to support users of a comprehensive online weight loss program.
Elucidating the Structural Evolution of O3-type NaNiFeMnO: A Prototype Cathode for Na-Ion Battery.
J Am Chem Soc
November 2024
State Key Laboratory of Physical Chemistry of Solid Surfaces, iChEM (Collaborative Innovation Center of Chemistry for Energy Materials), Department of Chemistry, College of Chemistry and Chemical Engineering, Xiamen University, Xiamen 361005, P. R. China.
Extending the depth-of-charge (DoC) of the layered oxide cathode presents an essential route to improve the competitiveness of the Na-ion battery versus the commercialized LiFePO-based Li-ion battery (0.8 CNY/Wh). However, the DoC-dependent boundary between detrimental/irreversible structural distortion and neutral/reversible structure interconversion cannot be clearly distinguished, which is attributed to the ambiguous recognition of correlation among the complex phase transition, local covalent environment evolution, and charge compensation.
View Article and Find Full Text PDFReducing semantic ambiguity in domain adaptive semantic segmentation via probabilistic prototypical pixel contrast.
Neural Netw
January 2025
School of Artificial Intelligence, Hebei University of Technology, No. 5340, Xiping Road, Tianjin, 300401, China.
Domain adaptation aims to reduce the model degradation on the target domain caused by the domain shift between the source and target domains. Although encouraging performance has been achieved by combining contrastive learning with the self-training paradigm, they suffer from ambiguous scenarios caused by scale, illumination, or overlapping when deploying deterministic embedding. To address these issues, we propose probabilistic prototypical pixel contrast (PPPC), a universal adaptation framework that models each pixel embedding as a probability via multivariate Gaussian distribution to fully exploit the uncertainty within them, eventually improving the representation quality of the model.
View Article and Find Full Text PDFHarnessing explainable artificial intelligence for patient-to-clinical-trial matching: A proof-of-concept pilot study using phase I oncology trials.
PLoS One
October 2024
School of Library and Information Studies, University of Oklahoma Norman Campus, Norman, Oklahoma, United States of America.
Article Synopsis
- The study develops an explainable AI system using Natural Language Processing (NLP) to match patients with phase 1 oncology clinical trials, aiming to improve recruitment efficiency for drug development.
- The prototype considers four key matching criteria: cancer type, performance status, genetic mutation, and measurable disease, and delivers a matching score with explanations of the results.
- Evaluations against expert matching revealed the system's performance metrics (73.68% precision, 56% recall, 77.78% accuracy, 89.36% specificity) and highlighted issues like abbreviation ambiguity and context misunderstanding as error factors, making it a unique approach in the public domain for this purpose.
Want 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!