Deep Learning-Based Emotion Detection

In order to make artificial intelligence smarter by detecting user emotions, this project analyzes and determines the current type of human emotions through computer vision, semantic recognition audio feature classification. facial expression recognition, for problems large number parameters poor real-time performance methods based on deep learning, Wang Weimin Tang Yang Z. et al. proposed a face method multilayer fusion with light-weight convolutional networks, which uses an improved inverted residual network as basic unit build lightweight model. Based method, experiment optimizes traditional CNN MobileNet model finally constructs new framework ms_model_M, has about 5% ms_model_M is tested two commonly used real datasets, FER-2013 AffectNet, accuracy 74.35% 56.67%, respectively, MovbliNet 74.11% 56.48% in tests these datasets. This structure well balances speed For emotion detection detection, existing models APIs are experiment.