"TIB AV Analytics" -- Development of a software platform for systematic film and video analysis

拟议的跨学科项目涉及媒体和计算机科学家，旨在开发一个基于网络的电影分析平台，支持媒体和电影研究以及系统电影和视频分析的其他领域。这包括开发用于自动视频分析的最先进的软件工具以及用于信息可视化的适应方法。拟议的平台“TIB AV Analytics”旨在扩展莱布尼茨科学技术信息中心（TIB）的AV门户网站。虽然目前的平台允许搜索科学视频，但不支持系统的科学视频和电影分析。因此，该项目将通过提供一个开放平台来支持依赖系统电影和视频分析的研究人员，该平台具有最先进的算法和特定的可视化，用于解释叙事结构。开发的图形用户交互软件组件、视频分析的新算法和可视化工具将进一步开源发布，从而提供给感兴趣的各方供后续使用。为了实现该项目，我们联合了具有从媒体科学角度进行系统电影分析以及自动视频分析和视觉分析专业知识的项目合作伙伴。电影学者贡献他们的领域知识的需求分析，开发平台的评估，并不断伴随着技术的发展。该计算机科学合作伙伴在视频分析软件Videana的开发中发挥了重要作用，并研究了多年的视频分析和视频检索系统。

The proposed interdisciplinary project involves media and computer scientists and aims at developing a web-based film analytics platform which supports media and film studies and other domains in systematic film and video analysis. This includes the development of state-of-the-art software tools for automatic video analysis as well as adapted approaches for information visualisation. The proposed platform "TIB AV Analytics" is supposed to extend the AV-Portal of the Leibniz Information Centre for Science and Technology (TIB). While the current platform allows for the search of scientific videos, systematic scientific video and film analysis are not supported. As a result, the project will support researchers who rely on systematic film and video analysis by offering an open platform featuring state-of-the-art algorithms and specific visualisations for the interpretation of narrative structures. The developed software components for graphical user interaction, novel algorithms for video analysis, and visualisation tools will furthermore be published open source and thus made available to interested parties for subsequent use. For the realisation of the project, we unite project partners with expertise in systematic film analysis from a media science perspective as well as automatic video analysis and visual analytics. The film scholars contribute with their domain knowledge to the requirements analysis, to the evaluation of the developed platform, and continuously accompany the technical development. The computer science partner accounted significantly for the development of the video analysis software Videana and researches video analysis and video retrieval system for several years.