%0 Journal Article %A JIANG Bin %A LI Nanxing %A ZHONG Rui %A WU Qinggang %A CHANG Huawen %T New Research Advances in Facial Expression Recognition Under Partial Occlusion %D 2022 %R 10.3778/j.issn.1002-8331.2112-0268 %J Computer Engineering and Applications %P 12-24 %V 58 %N 12 %X With the development of artificial intelligence technology, facial expression recognition can extract expression states from images or videos to recognize the mental emotions of human beings, achieving a better human-computer interaction effect. However, there are many methods only focusing on facial expression recognition without occlusion, which is not applicable to objective and complex scenarios and greatly limits the practicality of the algorithm. In recent years, aiming at different types of occlusion such as illumination occlusion, low resolution occlusion, pose variations, and physical occlusion, researchers have proposed various new methods to solve these problems. The survey introduces the core ideas of relevant methods, and conducts a comparison analysis about these methods, meanwhile looks forward to future research and development. %U http://cea.ceaj.org/EN/10.3778/j.issn.1002-8331.2112-0268