Automatic Generation of Descriptive Titles for Video Clips Using Deep Learning

Over the last decade, use of Deep Learning in many applications produced results that are comparable to and some cases surpassing human expert performance. The application domains include diagnosing diseases, finance, agriculture, search engines, robot vision, others. In this paper, we proposing an architecture utilizes image/video captioning methods Natural Language Processing systems generate a title concise abstract for video. Such system can potentially be utilized domains, including, cinema industry, video security surveillance, databases/warehouses, data centers, proposed functions operates as followed: it reads video; representative image frames identified selected; captioned; NLP is applied all generated captions together with text summarization; finally, All performed automatically. Preliminary provided paper using publicly available datasets. This not concerned about efficiency at execution time. We hope able address issues our subsequent publications.