Survey on Visual Rhythms: A Spatio-Temporal Representation for Video Sequences

Abstract Strategies for representing spatial and temporal information from video sequences in images are very convenient due to the demand for efficient tools for storing, browsing, retrieving, indexing and transmitting multimedia content. Visual rhythm, also known as temporal slice, summarizes relevant information present in a video through a space-time representation by means of one or a few images. In this work, we provide a comprehensive review of the main methods and techniques for constructing visual rhythms, classify the approaches into different categories, and identify new trends. Additionally, we discuss some limitations, challenges and applications related to the visual rhythm representation, as well as outline some directions for future research.