Short paper: Understanding the attention model of humans in sarcastic videos


Sarcasm is a common part of human communication that has long been ignored by sentiment analysis researchers. Sarcasm is also an important aspect in entertainment industry for TV series, movies etc. Recently, some works have shown the applicability of multimodality (e.g., image and text) in sarcasm research from a sentiment analysis perspective instead of text only approaches. However, none of those studies harness video. We argue videos can be interesting to study to understand the nature of sarcasm on social media. We study how sarcastic videos can gain an individual's attention and popularity at large. We show how an AI agent can suggest areas that might gain a viewer's attention in a sarcastic video. Identification of both attention gaining areas (AGA) and objects contained in sarcastic videos can be compared with the AGAs and objects in previously successful/popular sarcastic videos. In this paper, we present an AI agent to identify the optimal AGAs and one empirical study of objects commonly shown in directed sarcastic video settings.


Computer Science

Document Type

Conference Proceeding




Attention model, Sarcasm, Semantic segmentation, Videos

Publication Date