AI Image Analysis in Era of Short-time Viewing

Abstract

In the era when short videos are preferred, broadcasting stations have been enhancing momentums to distribute summary videos of broadcast content on social networking services (SNS). Therefore, we have developed automatic generation systems for news and programme summary videos. Using a video summarisation artificial intelligence (AI) that has learned the image composition and camerawork typical of important scenes, it is possible to automatically generate summary videos with a high quality close to videos edited by actual programme production staff. These systems have been on trial/practical use in various NHK broadcasting stations. The generated summary videos are posted daily on SNS. Furthermore, considering a programme website is also important content that could boost viewer contact rates, we developed a support system for programme website creation using an AI to extract thumbnails automatically. Using thumbnail candidate images extracted automatically by AI that has learned the unique features of a programme's representative images, you can create programme websites with minimal effort. These technologies can streamline the production of Internet-content such as summary videos and programme websites. Moreover, they will greatly boost Internet deployments of various broadcasting programmes.