Mixed reality is one of the most important techniques to achieve the vision of ubiquitous computing. Traditional middleware for mixed reality provide high level abstraction to hide complex algorithms for analyzing video images, but applications programmers still need to take into account distribution and automatic reconfiguration when developing mixed reality applications for ubiquitous computing.

Our middleware infrastructure hides all the complexities to build mixed reality applications for ubiquitous computing. Therefore, the development does not require advanced skills for ubiquitous computing. The paper describes the design and implementation of our infrastructure, and presents some scenarios and the current status showing its effectiveness.