Textual overlays are often used in social media videos as people who watch them without the sound would otherwise miss essential information conveyed in the audio stream. This is why extraction of those overlays can serve as an important meta-data source, e.g. for content classification or retrieval tasks. In this work, we present a robust method for extracting textual overlays from videos that builds up on multiple neural network architectures. The proposed solution relies on several processing steps: keyframe extraction, text detection and text recognition. The main component of our system, i.e. the text recognition module, is inspired by a convolutional recurrent neural network architecture and we improve its performance using synthetically generated dataset of over 600,000 images with text prepared by authors specifically for this task. We also develop a filtering method that reduces the amount of overlapping text phrases using Levenshtein distance and further boosts system's performance. The final accuracy of our solution reaches over 80% and is au pair with state-of-the-art methods.
Abstract. Considering the case in which not every individual is able to efficiently evacuate, due to the lack of knowledge about safe spots and available routes to them, close cooperation between community members plays a critical role. Using agent-based simulation, we tested two hypothetical scenarios of human behavior during tsunami evacuation and their efficiency, considering time and number of rescued people. In the first scenario, individuals did not cooperate with each other. In the second one, community members were trying to organize in groups, even if they could evacuate by their own. The results showed that in the second scenario not only substantially higher percentage of citizens evacuated in a shorter time, but it was nearly as efficient as the evacuation of a community in which almost every individual know at least one safe spot.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.