Facial action unit detection aims at making computers detect the action unit targets based on the given facial images or videos automatically. Due to a great amount of research during the past 20 years, especially the construction of more and more facial action unit databases and the raise of deep learning based methods, facial action unit detection technology has been rapidly developed. Firstly, the concept of facial action unit and commonly used facial action unit databases were introduced, and the traditional methods including steps such as pre-processing, feature extraction and classifier learning were summarized. Then, for several important research areas, such as region learning, facial action unit correlation learning and weak supervised learning, systematic review and analysis were conducted. Finally, the shortcomings of the existing reasearch and potential developing trends of facial action unit detection were discussed.

Aiming at the problem that the original on-line scoring system of Chinese films does not consider the information about the viewer who did not participate in the online survey, an improved scoring system based on evaluation participation rate was proposed. Firstly, an evaluation criterion of scoring system was established at the core of divergence and divergent effect based on the method of Regression Discontinuity Design (RDD). Secondly, distinguished from the weighted average method, an improvement method using the participation rate was put forward. Lastly, empirical study was conducted on the films released from 2014 to 2016 in China. The results show that the improved score's change rate of divergent effects after normalization of deviation is less than or approximate to 0, while the change rate of weighted average score is nearly 40%. Therefore, it is reasonable and feasible to analyze the differences between different scoring systems by using the divergent point and divergent effects; the improved score is closer to the real reputation influence of the film with less divergent effect after normalization, and can more intuitively reflect the sorting position of a film.