Transformative technology and demonstrator for immersive and naturalistic video communication

众所周知，在与他人交流时，我们分享的信息中只有50%来自语言和语气，其余的来自视觉线索，如凝视，肢体语言和我们的表情。今天的视频通信在捕获和传达这些视觉线索方面明显不足，因此当被认为是面对面会议的替代方案时，特别是对于更重要和关键的通信，通常福尔斯。这通常是由于视频通信的非真实感和视角差。该项目涉及研究，探索和开发一种技术，通过执行视频流的实时操作，帮助捕获和传达这些视觉提示，从而显著改善视频通信的体验。该项目旨在以大众市场的方式实现这一目标，并对所有视频通信用户产生有意义的影响。这使得更多的面对面互动过渡到数字互动，从而减少了不必要的旅行需求，对全球碳减排目标产生积极影响，并帮助更多的人实现更大的经济影响，而不必在人口过剩的城市中共处。这将为英国和国际社会带来巨大的经济效益。该项目的大部分工作集中在深度技术和算法开发上，推动基于机器学习的实时信号处理的最新技术。

It is well understood that when communicating with others, only 50% of the information we share lies in the words and our tone, and the rest in visual cues such as gaze, body language and our expressions. Video communication today falls significantly short in capturing and communicating these visual cues and hence typically falls short when considered as an alternative to meeting face-to-face, especially for more important and critical communications. This is often due to the non-real feeling of video communication, and poor perspective of view.The project involves research, exploration and development of a technology that enables significant improvements to the experience of video communication, by performing real-time manipulation of video streams, that helps capture and communicate these visual cues. The project aims to realise this in a mass market way and have a meaningful impact to all users of video communication. This enables the transition of more face-to-face interactions to digital ones, and in turn reduce the need for unnecessary travel, positively impacting global carbon reduction targets, and help more people achieve greater economic impact without having to co-locate in overpopulated cities. This will be a significant economic benefit to the UK and internationally.The majority of the project work focuses on deep technology and algorithmic development, pushing the state-of-the-art in machine learning based real-time signal processing.