20.500.12556/RUL-176579 EmoVisioNet a hybrid network unifying lightweight CNN and attention-based vision model for facial emotion detection Facial emotion detection has witnessed a surge in demand across numerous applications, including human-computer interaction, healthcare, and security. Accurate expression recognition is crucial for improving human-computer interactions and understanding human behavior. Existing facial emotion detection models face challenges in achieving both high accuracy and real-time processing due to complex architectures. Our goal is to create an efficient yet accurate solution that can work on resource-constrained devices. To address the challenge of accurately recognizing emotions from facial expressions, we propose a novel hybrid approach that combines the strengths of pretrained Lightweight Convolutional Neural Networks (CNNs), and Attention-based Vision Models. The pretrained Lightweight CNN serves as a feature extractor, efficiently capturing facial features, while the attention model refines the feature representation to focus on crucial regions of the face associated with different expressions. This enables our model to achieve state-of-the-art (SOTA) accuracy with reduced computational requirements. The proposed model, EmoVisioNet, achieves superior performance across multiple datasets, attaining 99.97 % accuracy on CK+, 96.23 % on RAF-DB, 93.88 % on FER2013, and 96.91 % on FERPlus. The obtained results surpass the current state-of-the-art in this field, demonstrating EmoVisioNet’s superior performance in facial expression recognition. artificial intelligence computer vision machine learning facial analysis facial expression recognition attention-based vision network convolutional neural network emotional intelligence umetna inteligenca računalniški vid strojno učenje analiza obrazov true false true Angleški jezik Slovenski jezik Članek v reviji 2025-12-04 11:22:34 2025-12-04 11:22:39 2026-07-07 04:17:35 0000-00-00 00:00:00 2026 0 0 12 str. art. 132224 Vol. 665 7 Feb. 2026 0000-00-00 Zaloznikova Objavljeno NiDoloceno 0000-00-00 0000-00-00 0000-00-00 004.8 1872-8286 10.1016/j.neucom.2025.132224 259745027 1-s2.0-S0925231225028966-main.pdf 1-s2.0-S0925231225028966-main.pdf 1 1D404EA32D54F49A969CC17B60D1D743 958e01effe442dd9ed263b9e1aff116eaa72ebef63cc63313b6abbbdb553d2fa b272a0ca-792a-11f1-9b0d-0050569b8976 https://repozitorij.uni-lj.si/Dokument.php?lang=slv&id=237124 https://www.sciencedirect.com/science/article/pii/S0925231225028966 0 2631e980-d0fb-11f0-9328-0050569b8976 https://repozitorij.uni-lj.si/Dokument.php?lang=slv&id=222998 Fakulteta za elektrotehniko 0 0 0