Facial Expression Recognition Based on the Combination of Landmarks and Texture Features
Weiyang Chen, Ke Wu, Yi Feng Pan · 2023
Facial expression recognition plays an important role in understanding people’s emotion and social interaction. It is meaningful to design method to recognize facial expression automatically. It can be applied to many situations, such as emotional state recognition, monitoring the status of students in class, monitoring the status of drivers, online recommendation system and so on. This paper proposes a facial expression recognition method based on the combination of landmarks and local texture patterns. First, find out the Landmarks. Next, the facial components are extracted from the facial image. The local texture feature is extracted from facial components. In order to combine these features for facial expression recognition, we have carried out data normalization, redundancy removal, and dimension reduction. The experimental results on the JAFFE dataset and CK+dataset show that the proposed facial expression recognition method based on the combination of landmarks and the local texture patterns is effective.