The analysis of dining events is important and useful for a wide range of applications such as smart restaurants, and for different research areas like sociologists' studies and social interactions analysis. Particularly, the performance and reputation of restaurants is completely dependent on customers satisfaction as a possible metric. With the rapid growth of computer vision technologies, smart restaurants are going to leverage such technologies for indirectly measuring and quantifying customer satisfactions through analyzing videos and images without performing any direct questioner process. However, the large volume of recorded data by cameras makes the manual analysis process computationally expensive in terms of time to quantify customer satisfaction. Hence, in this paper, we introduce a design of a framework, so-called DiEvent, that integrates various components for automatically analyzing dining events. Our framework could be leveraged in different applications and researches, including cooking recipe evaluation in terms of customer satisfaction, performing sociology studies in dining events, and social interactions detection researches.