Facial emotion recognition is an important topic of study in artificial intelligence and computer vision. In this paper, we offer a novel automated system for facial emotion identification that uses fusion approaches and a novel model architecture based on different flavors of MobileNet. MobileNet is widely used as a lightweight deep learning model; nonetheless, its application fusion technique with a range of its own flavors is novel. This paper introduces \(M^3\) SI-Net, a novel facial emotion recognition model that leverages a fusion approach based on variants of MobileNet (MobileNetV2, MobileNetV3Small, and MobileNetV3Large). The model integrates Inception blocks and a re-parameterized Swish1 function to enhance feature extraction and classification accuracy. Through extensive experiments on the Jaffe and Cohn-Kanade datasets, we demonstrate that \(M^{3}\) SI-Net achieves state-of-the-art performance, outperforming existing methods in terms of accuracy, precision, recall, and F1-score. Our approach provides new insights into facial emotion recognition and showcases the potential of fusion models in improving the accuracy and robustness of deep learning-based systems. This research looks at the importance of face expression recognition, which is important in human-computer interaction, affective computing, and a variety of applications in healthcare, entertainment, and security. Because of the intrinsic complexity and diversity of human face emotions, we needed to come up with creative solutions, which led us to use siameze learning approaches. The Siameze multiflavoured mobilenet model for a data fusion-based technique is a fundamental contribution of this work, boosting its expressiveness for greater adaptability. Our technique effectively leverages different information by merging multiple base models with various pre-processings, resulting in higher facial emotion identification accuracy. The combination of complimentary data from diverse sources improves the system’s accuracy and robustness. Our code will be made publicly available through this https://github.com/stuj2019/M-3SI.
