Article: Eye movement fixation point localisation and recognition method based on multimodal features Journal: International Journal of Data Mining and Bioinformatics (IJDMB) 2026 Vol.30 No.7 pp.108 - 132 Abstract: This study proposes an advanced framework for eye fixation localisation and recognition by integrating the multimodal eye fixation network (MEFN) with an adaptive multimodal integration strategy (AMIS). MEFN combines visual, temporal, and contextual information through convolutional neural networks, recurrent neural networks, and attention mechanisms, enabling the model to capture complex spatial-temporal dependencies in eye movement data. AMIS further improves the framework by dynamically adjusting the contribution of each modality according to contextual relevance, signal quality, and data reliability. A reinforcement learning feedback loop is introduced to support real-time optimisation and continuous adaptation under different experimental conditions. Experiments conducted on benchmark datasets demonstrate that the proposed framework achieves higher accuracy, stronger robustness, and better computational efficiency than existing eye fixation localisation and recognition methods. Ablation studies confirm the effectiveness of each component. the framework shows promising potential for cognitive research, human-computer interaction, assistive technology, and medical diagnostic applications. Inderscience Publishers - linking academia, business and industry through research

Title: Eye movement fixation point localisation and recognition method based on multimodal features

Authors: Yang Tong; Shaoyun Bai

Addresses: Shanxi University of Electronic Science and Technology, Linfen, 041000, China ' Shanxi University of Electronic Science and Technology, Linfen, 041000, China

Abstract: This study proposes an advanced framework for eye fixation localisation and recognition by integrating the multimodal eye fixation network (MEFN) with an adaptive multimodal integration strategy (AMIS). MEFN combines visual, temporal, and contextual information through convolutional neural networks, recurrent neural networks, and attention mechanisms, enabling the model to capture complex spatial-temporal dependencies in eye movement data. AMIS further improves the framework by dynamically adjusting the contribution of each modality according to contextual relevance, signal quality, and data reliability. A reinforcement learning feedback loop is introduced to support real-time optimisation and continuous adaptation under different experimental conditions. Experiments conducted on benchmark datasets demonstrate that the proposed framework achieves higher accuracy, stronger robustness, and better computational efficiency than existing eye fixation localisation and recognition methods. Ablation studies confirm the effectiveness of each component. the framework shows promising potential for cognitive research, human-computer interaction, assistive technology, and medical diagnostic applications.

Keywords: multimodal integration; eye movement analysis; fixation point localisation; neural networks; attention mechanism; adaptive fusion.

DOI: 10.1504/IJDMB.2026.154770

International Journal of Data Mining and Bioinformatics, 2026 Vol.30 No.7, pp.108 - 132

Received: 16 Mar 2026
Accepted: 15 May 2026
Published online: 13 Jul 2026 *

Title: Eye movement fixation point localisation and recognition method based on multimodal features

Keep up-to-date