Semantic Media: a new paradigm for navigable content for the 21st Century

语义媒体：21 世纪可导航内容的新范式

• 批准号: EP/J010375/1
• 负责人: Mark Sandler
• 金额: $ 72.98万
• 依托单位: Queen Mary University of London
• 依托单位国家: 英国
• 项目类别: Research Grant
• 财政年份: 2012
• 资助国家: 英国
• 起止时间: 2012 至 无数据
• 项目状态: 已结题
• 来源: https://gtr.ukri.org/projects?ref=EP%2FJ010375%2F1
• 关键词: Semantic; Media; new; paradigm; navigable

This proposal stems directly from the EPSRC Workshop held on 20 & 21 October 2010 "ICT Research - The Next Decade". It seeks to address the challenge of the navigation of time-based media collections and items throughout the content life-cycle, from creation to consumption. It will achieve this by establishing an open network of researchers from across academia and industry, who engage in workshops, sandpits and, most importantly, feasibility or path-finder mini-projects. These mini-projects have the aims not only of performing leading edge, early stage research that will lead on to larger proposals, but also of building a critical mass of researchers, whose expectation is to tackle significant challenges by collaborating. Other elements of this project are to create a Landscape document for the field, develop appropriate ontologies for capturing media semantics, present results through a diverse range of channels and summarise the findings of the project, including a Roadmap. The research agenda is based on five premises: 1. Content-related metadata is an effective and scalable approach exemplified in this domain and applicable to future large scale, automated and interactive information systems;2. The point of creation is the best time and place to collect (and compute) metadata; 3. The best way to represent this metadata is one that is amenable to knowledge processing and management, linked data strategies and logical inference;4. Significant challenges require a cross-disciplinary approach, ranging from fundamental theory to applied research set in the context of a real problem; and 5. The UK is supremely placed with the world-leading skills and experience to be a world-beating authority in an area of intellectual and societal/commercial benefit.This proposal deliberately does not address related problems of navigation through legacy content, nor of Rights Management, as these are already embedded in the research landscape. Instead, we concentrate on the production of future media items.The issues raised and investigated by this proposal are pertinent not only to EPSRC, but also to ESRC, AHRC, JISC and TSB, and with particular relevance to the Digital Economy.The applications of such ideas span all the different time-based media, including music, drama, documentary, film, texts and so on. In order to advance the field, this project will bring together acknowledged experts from across UK academia in a diverse range of disciplines, including Semantic Web experts, Signal Processing experts, Video experts, Performance experts and more. The project aims to form a network and a critical mass of expertise by a series of interventions that will also include industrial collaborators (assisted by the TSB Creative Industries Knowledge Transfer Network). Network activities include workshops and sandpits, as well as collaborative small scale research projects, each typically of 6 months duration with 2 or 3 participant universities.The outcomes of the project include: Research and Impact Roadmaps; a well-connected community of researchers engineers, creatives, content producers and funders; commercial and full-fledged research proposals; research publications; and specific impact activities at world leading Broadcast and Media conventions.

这一建议直接源于EPSRC于2010年10月20日和21日举办的题为“信息和通信技术研究--下一个十年”的研讨会。它力求解决从创作到消费的整个内容生命周期中基于时间的媒体收藏和项目导航的挑战。它将通过建立一个由来自学术界和工业界的研究人员组成的开放网络来实现这一点，这些研究人员参与研讨会、沙坑，最重要的是可行性或探路者微型项目。这些微型项目的目标不仅是进行前沿的早期研究，这些研究将导致更大的提案，而且还旨在建立一支关键的研究人员队伍，他们的期望是通过合作来应对重大挑战。该项目的其他内容是为实地创建一份景观文件，开发适当的本体论来捕捉媒体语义，通过各种渠道展示结果，并总结项目的结果，包括路线图。研究议程基于五个前提：1.与内容相关的元数据是这一领域的范例，是一种有效和可扩展的方法，适用于未来的大规模、自动化和交互式信息系统；2.创建点是收集(和计算)元数据的最佳时间和地点；3.表示这种元数据的最佳方式是服从知识处理和管理、关联数据策略和逻辑推理的方式；4.重大挑战需要跨学科的方法，从基础理论到实际问题背景下的应用研究；5.英国拥有世界领先的技能和经验，在知识和社会/商业利益领域成为世界一流的权威机构。这项建议故意没有解决导航传统内容或权利管理的相关问题，因为这些问题已经嵌入研究版图。这项建议所提出和调查的问题，不单与EPSRC有关，还与ESRC、AHRC、JISC和TSB有关，尤其与数码经济有关。这些概念的应用跨越所有不同的以时间为基础的媒体，包括音乐、戏剧、纪录片、电影和文本等。为了推动该领域的发展，该项目将汇集来自英国学术界不同学科的公认专家，包括语义网专家、信号处理专家、视频专家、性能专家等。该项目旨在通过一系列干预措施形成一个网络和一大批专业知识，其中还将包括工业合作者(在TSB创意产业知识转移网络的协助下)。网络活动包括讲习班和沙坑，以及合作的小规模研究项目，每个项目通常为期6个月，由2到3所参与大学参与。该项目的成果包括：研究和影响路线图；研究人员、工程师、创意人员、内容制作人和资助者的良好联系社区；商业和成熟的研究提案；研究出版物；以及在世界领先的广播和媒体大会上开展的具体影响活动。

项目成果

Improving Time-Scale Modification of Music Signals Using Harmonic-Percussive Separation

• DOI: 10.1109/lsp.2013.2294023
• 发表时间: 2014
• 期刊: IEEE Signal Processing Letters
• 影响因子: 3.9
• 作者: Jonathan Driedger;Meinard Müller;Sebastian Ewert

Recent advances in affective and semantic media applications at the BBC

BBC 在情感和语义媒体应用方面的最新进展

• DOI: 10.1109/wiamis.2013.6616134
• 发表时间: 2013
• 作者: Eggink J

Explorations in Linked Data practice for early music corpora

早期音乐语料库关联数据实践探索

• DOI: 10.1109/jcdl.2014.6970184
• 发表时间: 2014
• 作者: Crawford T

Event-driven retrieval in collaborative photo collections

协作照片集中的事件驱动检索

• DOI: 10.1109/wiamis.2013.6616121
• 发表时间: 2013
• 作者: Brenner M

Score-informed audio decomposition and applications

• DOI: 10.1145/2502081.2502143
• 发表时间: 2013-10
• 期刊: Proceedings of the 21st ACM international conference on Multimedia
• 作者: Jonathan Driedger;H. Grohganz;Thomas Prätzlich;Sebastian Ewert;Meinard Müller