Micro-temporal facial expressions consist of brief and subtle muscle activations that convey genuine emotional states. Accurately modeling these transient dynamics is challenging due to their low intensity, short duration, and limited spatial variation, which often hinder the performance of conventional convolutional and recurrent architectures. To address these limitations, this paper introduces a Spatio-Temporal Self-Attention Network (STTN) that leverages transformer-based self-attention to effectively capture finegrained dependencies across both spatial regions and short temporal intervals. The proposed framework focuses on learning discriminative representations of micro-temporal facial movements by emphasizing relevant facial regions and their temporal evolution. Extensive experiments are conducted on high-frame-rate micro-expression benchmarks, including SAMM, CASME II, and CAS(ME)2 datasets. The results demonstrate that the proposed model achieves superior performance compared to existing state-of-the-art approaches, highlighting the effectiveness of self-attention mechanisms in modeling subtle and rapid facial dynamics.
J. R, K. Sindhubala, D. Kiruba et al.· International Conference on...· 0 citations
The early detection of individuals with Mild Cognitive Impairment (MCI) who are at high risk of developing Alzheimer’s Disease (AD) is important for proper clinical intervention and disease management. Studies show that there is a strong link between brain age, which is estimated from brain scan data, and actual chronological age. This difference is called the Brain Age Gap (BAG), and it acts as a useful biological marker for identifying people at risk of brain degeneration. This study aims to create an automated dual-network framework that predicts MCI subjects’ likelihood of converting to AD based on their structural magnetic resonance imaging (MRI) brain scans. The two networks will consist of (i) a three-dimensional convolutional neural network (CNN) model that predicts an estimated brain age based on the inputted MRI scan of a subject, and (ii) a risk prediction network that predicts probability of MCI-to-AD conversion by incorporating estimated brain age, BAG, and deep feature representations. The framework will be trained in sequential order using a longitudinal neuroimaging dataset. Results from the experiment indicate higher classification performance can be attained with the proposed dual-network architecture over current state-of-the-art single-stage classification methods or models without brain age. Furthermore, these results strongly support the use of brain age-based predictive features for early prediction of AD risk. The proposed method offers a fully automated and clinically interpretable solution for supporting early diagnosis and personalized intervention planning in Alzheimer’s disease.
B. Stanley, K. Sindhubala, J. S. Shemona et al.· International Conference on...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
