RORMIX - A Music Video Technology Using Novel Active Machine Learning to embed image and video on to tagged assets in other music videos.

RORMIX - 一种音乐视频技术，使用新颖的主动机器学习将图像和视频嵌入到其他音乐视频中的标记资产中。

• 批准号: 131991
• 金额: $ 2.94万
• 依托单位: RORMIX LIMITED
• 依托单位国家: 英国
• 项目类别: Feasibility Studies
• 财政年份: 2015
• 资助国家: 英国
• 起止时间: 2015 至 无数据
• 项目状态: 已结题
• 来源: https://gtr.ukri.org/projects?ref=131991
• 关键词: RORMIX; Music; Video; Technology; Using

RORMIX connects music video's of emerging/unsigned artists to millions of personal/professional consumers. Their imaging tech enables major brands to embed logo or other brand advertising messages into existing video. Currently a labour intensive manual process. This project aims to prove feasibility of integrating a machine learning algorithm capable of (1) identifying target digital assets in video (such as a flat wall/floor surface, etc) and (2) automatically embedding an image or other video onto that tagged location. Success enables RORMIX to operate a virtual 'try-before-you-buy' business model, allowing clients to remotely select target assets in a music video, and examine how their brand appears, before committing. RORMIX is currently downloaded in >100 countries. Key challenges: (1) recognition/ tagging, appropriate target assets in video; (2) embedding; (3) enabling over mobile platform via advanced user-interface. Market: 40% of YouTube traffic comes from mobile (Gartner, 2014). 50mn US citizens watch video on mobiles. Global Music on Demand market is $197bnUSD, CAGR 20% (PR NewsWire).

RORMIX将新兴/未签约艺术家的音乐视频与数百万个人/专业消费者联系起来。他们的成像技术使主要品牌能够将徽标或其他品牌广告信息嵌入现有视频中。目前是一个劳动密集型的手工过程。该项目旨在证明集成机器学习算法的可行性，该算法能够（1）识别视频中的目标数字资产（例如平坦的墙壁/地板表面等），以及（2）自动将图像或其他视频嵌入到标记的位置上。成功使RORMIX能够运营虚拟的“先试后买”的商业模式，允许客户在音乐视频中远程选择目标资产，并在承诺之前检查他们的品牌如何出现。RORMIX目前在超过100个国家下载。关键挑战：（1）识别/标记视频中的适当目标资产;（2）嵌入;（3）通过高级用户界面在移动的平台上启用。市场：40%的YouTube流量来自移动的（Gartner，2014）。5000万美国人在手机上观看视频全球音乐点播市场是1970亿美元，复合年增长率20%（公关新闻网）。