课题基金 / 基金详情

Novel AI video content moderation system

Novel AI video content moderation system
新颖的人工智能视频内容审核系统
批准号:
10029108
负责人:
金额:
$48.17万
依托单位国家:
英国
项目类别:
Collaborative R&D
财政年份:
2022
资助国家:
英国
项目状态:
已结题
起止时间:
2022 至 --

项目摘要

项目成果

相似基金

相关文献

中文摘要
翻译
理解、文化相关、值得信赖的年龄评级对于媒体内容受众做出知情的观看选择至关重要。在英国,立法确保电影和包装媒体(DVD/蓝光)内容由英国电影分类委员会(BBFC)进行分类,但对于OTT内容的此类规定没有法定要求,尽管92%的英国公众认为它们显示与电影和DVD/蓝光相同的年龄等级很重要(BBFC,2020)。OTT现在已经成为主流,从以前发布内容的聚合器发展到自己的创建者,今天比传统媒体提供商生产更多的媒体内容。53%的英国家庭订阅1+ OTT(2020年)。96%的5-15岁奥尔兹观看OTT内容,98%的父母主动限制/调解他们的孩子使用OTT,这表明越来越需要将高效和可信的分类系统扩展到在线媒体(Ofcom,2020)。OTT的增长速度让世界各地的立法者迎头赶上。在立法真空的情况下,一些OTT服务(如Netflix)寻求与BBFC等监管机构合作,为其用户提供100%的可信、与文化相关的年龄评级,其他人则暂时选择了不同的道路。使OTT对消费者如此有吸引力的媒体内容数量是年龄评级分配的挑战。传统的分类方法,其中一个人的看法,在每个领域的所有内容完全是不可行的财政或后勤。这导致部分OTT提倡自我监管及自行分配年龄评级或向第三方寻求评级,导致OTT使用不获认可或与文化相关的不符合标准的年龄评级,令非成年人的心理健康面临风险。然而,很明显,即使在OTT年龄评级立法的地区(新西兰/澳大利亚/德国/法国),也需要一个可行的、全球性的、可扩展的、具有成本效益的分类解决方案。BBFC开发了一种正在申请专利的工具(2115955.3),可以通过解释详细的分类数据为多个地区分配准确的本地化年龄评级。在BBFC的领导下,该创新英国项目将使用合格财团的工具,数据,资源和专业知识(包括巴斯大学和Inventya)和分包商技术合作伙伴Amazon Web Services UK,开发能够标记分类问题的人工智能,以驱动“自动分类产品(ACP)”,为OTT提供低成本,为多个地区提供高质量年龄评级的内部解决方案-保护其用户和品牌。
英文摘要
Understood, culturally relevant, trustworthy age-ratings are vital for media content audiences to enable informed viewing choices. In the UK, legislation ensures that cinema and packaged media (DVD/Blu-Ray) content are classified by the British-Board-of-Film-Classification(BBFC), however there is no statutory requirement for such provisions for Over-The-Top(OTT) content despite 92% of the UK public believing it important for them to display the same age-ratings as cinema and DVD/Blu-Ray (BBFC, 2020). OTTs are now mainstream, having evolved from aggregators of previously-released content, to creators of it themselves, producing more media content today than traditional media providers. 53% of UK households subscribe to 1+ OTT (2020). 96% of 5-15 year olds watch OTT content, with 98% of parents proactively restricting/mediating their child's use of OTT, indicating a growing need to extend efficient and trusted classification systems to online media (Ofcom,2020).The speed at which OTTs have grown has left legislators around the World playing catch-up. In the legislative vacuum some OTT services, such as Netflix, have sought to work with regulators like the BBFC to provide their users with 100% coverage of trustworthy, culturally-relevant age ratings, others have temporarily chosen a different path. The volume of media content that makes OTTs so attractive to consumers, is a challenge for age-rating assignment. Traditional methods of classification where a human views all content in full in each territory is not viable financially or logistically. This has led some OTTs to champion self-regulation and self-assign age-ratings or seek them from a third-party, leading to a wild-west situation where the mental-health of young-people is put at risk as substandard age-ratings that are not recognised or culturally relevant are used on OTTs.Concerned, DCMS(UKGov) launched a consultation in 2021 titled 'Audience protection standards on Video-on-Demand(VoD) Services'. However, it's clear even in territories with OTT age-rating legislation (New Zealand/Australia/Germany/France), a viable, global, scalable, cost-effective classification solution is needed. The BBFC have developed a patent-pending(2115955.3) tool that can assign accurate, localised-age-ratings to multiple territories by interpreting detailed classification data. Led by BBFC, this Innovate-UK-project will use tools, data, resources, and expertise of a qualified consortium (including University of Bath, and Inventya) and subcontractor technology partner Amazon Web Services UK, to develop AI capable of tagging classification issues to drive an 'Automated Classification Product (ACP)' providing OTTs with a low-cost, in-house solution that delivers high-quality age-ratings for multiple-territories - protecting their users and brand.
期刊论文(0)
专著(0)
科研奖励(0)
会议论文
国内基金
海外基金
面向AI驱动的信息化工程监管与自动化测试平台研发
  • 批准号:
  • 项目类别:
    省市级项目
  • 资助金额:
    --
  • 批准年份:
    2026
  • 负责人:
    刘登志
  • 依托单位:
建筑-音乐跨模态AI生成平台研发与应用
  • 批准号:
  • 项目类别:
    省市级项目
  • 资助金额:
    --
  • 批准年份:
    2026
  • 负责人:
    许蕴彰
  • 依托单位:
适用于AI眼镜的横向错位光学变焦系统技术开发
  • 批准号:
  • 项目类别:
    省市级项目
  • 资助金额:
    --
  • 批准年份:
    2026
  • 负责人:
    窦健泰
  • 依托单位:
AI赋能中国传统壁画大模型开发与数字再生展示
  • 批准号:
  • 项目类别:
    省市级项目
  • 资助金额:
    --
  • 批准年份:
    2026
  • 负责人:
    朱亮亮
  • 依托单位: