SBIR Phase I: Facial Animation for Communication Enhancement

这个来自Seamless Solutions， Inc.的小企业创新研究第一阶段项目使用签字人的头部、面部表情和嘴巴的虚拟现实表示来增强口头和/或美国手语（ASL）交流，它提供了一种使用命令来动画这些模型的方法，这些命令可以通过廉价、低带宽的渠道传输。该项目的目标是建立在无缝解决方案目前的研究和开发，通过虚拟的人类代表或化身进行远距离手语交流。具体而言，该项目将探索使用低成本计算机生成的面部动画的可行性：1)表达情感，2)实现唇读，以及3)通过头部运动来增强肢体动作，从而增加交流的重点。这项技术不仅可以提供低成本、远距离的手语交流，还可以在人与计算机应用程序之间提供更自然、更快速的交互。将从这项研究中受益的人包括50多万听力残疾的美国人，他们目前无法通过长途通信系统用他们最具表现力的习语进行交流。此外，为聋哑人加强视觉交流的技术也可以为听力正常的人增加交流的表达能力。面部动画可以授权给教育、通信和娱乐软件开发商（包括游戏开发商、基于pc的软件开发商、数字电视开发商和可视电话开发商）。此类许可将导致在1)教育多媒体软件中使用虚拟形象；2)电脑游戏；3)电视节目；4)互动交流，这是今天声音被利用的地方。当与语音识别、自然语言处理和面部表情识别等新兴技术相结合时，这些虚拟形象可以添加到现有媒体中，而无需独立创作他们的行为。

This Small Business Innovation Research Phase I project from Seamless Solutions, Inc. uses virtual reality representations of a signer's head, facial expressions, and mouth to augment verbal and/or American Sign Language (ASL) communication, and it provides a means to animate these models using commands which can be transmitted over inexpensive, low bandwidth channels. The objective of this project is to build on Seamless Solutions' current research and development in the long distance communication of sign language by virtual human representations or avatars. Specifically, the project will explore the feasibility of using low-cost computer generated facial animations 1) to express emotion, 2) to enable lip reading, and 3) to augment signing with head movements that add emphasis to communication. This technology will not only provide low cost, long-distance sign language communication, but will also supply more natural and rapid interaction between humans and computer applications. Among those who will benefit form this research are the more than 500,000 hearing disabled Americans who are currently excluded from communicating in their most expressive idiom over long distance communications systems. Moreover, the same technologies that will enhance visual communication for the deaf can also add expression to communications for persons with unimpaired hearing. Facial animations can be licensed to education, communications, and entertainment software developers (including developers of games, PC-based software, digital TV, and videophones). Such licensing would result in avatar use in 1) educational multimedia software; 2) computer games; 3) television programs; and 4) interact communications, where sound is utilized today. When coupled with emerging technologies in voice recognition, natural language processing, and facial expression recognition, the avatars can be added to existing media without the cost of independently authoring their behavior.