AI Article Synopsis
- Three-dimensional face recognition plays a significant role in computer vision, with point clouds being a common representation despite their unordered nature and lack of texture detail affecting feature recognition.
- This paper introduces a novel face recognition network comprising a local feature descriptor for fine-grained feature extraction and a permutation-invariant approach to address the unordered point clouds problem.
- Experimental results on three datasets (CASIA-3D, Bosphorus, and Lock3Dface) demonstrate that this new method enhances accuracy by 0.7%, 0.4%, and 0.8% compared to existing techniques.
Article Abstract
Three-dimensional face recognition is an important part of the field of computer vision. Point clouds are widely used in the field of 3D vision due to the simple mathematical expression. However, the disorder of the points makes it difficult for them to have ordered indexes in convolutional neural networks. In addition, the point clouds lack detailed textures, which makes the facial features easily affected by expression or head pose changes. To solve the above problems, this paper constructs a new face recognition network, which mainly consists of two parts. The first part is a novel operator based on a local feature descriptor to realize the fine-grained features extraction and the permutation invariance of point clouds. The second part is a feature enhancement mechanism to enhance the discrimination of facial features. In order to verify the performance of our method, we conducted experiments on three public datasets: CASIA-3D, Bosphorus, and Lock3Dface. The results show that the accuracy of our method is improved by 0.7%, 0.4%, and 0.8% compared with the latest methods on these three datasets, respectively.
Download full-text PDF |
Source |
|---|---|
| http://www.ncbi.nlm.nih.gov/pmc/articles/PMC10537083 | PMC |
| http://dx.doi.org/10.3390/s23187715 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Visual short-term memory binding and attentional processes during object-scene integration are preserved in mild cognitive impairment.
Cortex
December 2024
Department of Psychology, Sapienza University of Rome, Rome, Italy; IRCCS Fondazione Santa Lucia, Rome, Italy. Electronic address:
Binding, a critical cognitive process likely mediated by attention, is essential for creating coherent object representations within a scene. This process is vulnerable in individuals with dementia, who exhibit deficits in visual working memory (VWM) binding, primarily tested using abstract arrays of standalone objects. To explore how binding operates in more realistic settings across the lifespan, we examined the impact of object saliency and semantic consistency on VWM binding and the role of overt attention.
View Article and Find Full Text PDFDevelopment and validation of a schizophrenia literacy scale for individuals diagnosed with schizophrenia.
Psychiatry Res
December 2024
School of Nursing, College of Medicine, Chang Gung University, Tao-Yuan City, Taiwan; Department of Nursing, Chang Gung University of Science and Technology, Tao-Yuan City, Taiwan; Department of Psychiatry, Chang Gung Memorial Hospital, Keelung Branch, Keelung City, Taiwan. Electronic address:
Schizophrenia literacy is key to early identification of schizophrenia and diminishment of stigma. This facilitates help-seeking and adherence to mental health treatment for individuals suffering from schizophrenia. However, validated measures assessing schizophrenia literacy among individuals diagnosed with schizophrenia remain limited.
View Article and Find Full Text PDFDomain adaptation spatial feature perception neural network for cross-subject EEG emotion recognition.
Front Hum Neurosci
December 2024
School of Computer Sciences, Universiti Sains Malaysia, Penang, Malaysia.
Emotion recognition is a critical research topic within affective computing, with potential applications across various domains. Currently, EEG-based emotion recognition, utilizing deep learning frameworks, has been effectively applied and achieved commendable performance. However, existing deep learning-based models face challenges in capturing both the spatial activity features and spatial topology features of EEG signals simultaneously.
View Article and Find Full Text PDFPerceptions of Resilience Among Nurses: A Qualitative Study Based on the Society-To-Cells Framework.
J Adv Nurs
December 2024
Department of Community Health Nursing, College of Nursing, Jouf University, Sakaka, Saudi Arabia.
Aim(s): To explore the perceptions of resilience among nurses using the Society-to-Cells Resilience Theory and examine how multilevel factors influence their ability to maintain resilience in high-stress environments.
Design: A qualitative study using semi-structured interviews.
Methods: Sixteen registered nurses from various healthcare settings in the Asir region, Saudi Arabia, participated in face-to-face interviews conducted from February to April 2024.
Neuropsychological profile of POLR3A-related spastic ataxia.
Neurol Sci
December 2024
Memory Clinic, Department of Neurology, Onze-Lieve-Vrouwziekenhuis, Aalst, Belgium.
Background And Objectives: POLR3-related disorders are a group of autosomal recessive neurodegenerative diseases that usually cause leukodystrophy and can lead to cognitive dysfunction. Literature reporting comprehensive neuropsychological assessment in POLR3A-related diseases is sparse. Here we describe the neuropsychological profile of a case of childhood-onset POLR3A-related spastic ataxia without leukodystrophy.
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!