Man-machine interaction is the gateway providing access to functions and services, which, due to the ever increasing complexity of smart systems, threatens to become a bottleneck. This book therefore introduces not only advanced interfacing concepts, but also gives insight into the related theoretical background.This refers mainly to the realization of video-based multimodal interaction via gesture, mimics, and speech, but also to interacting with virtual object in virtual environments, cooperating with local or remote robots, and user assistance.

While most publications in the field of human factors engineering focus on interface design, this book puts special emphasis on implementation aspects. To this end it is accompanied by software development environments for image processing, classification, and virtual environment implementation. In addition a test data base is included for gestures, head pose, facial expressions, full-body person recognition, and people tracking. These data are used for the examples throughout the book, but are also meant to encourage the reader to start experimentation on his own.

Thus the book may serve as a self-contained introduction both for researchers and developers of man-machine interfaces. It may also be used for graduate-level university courses.

Zusammenfassung

Contemporary man-machine interfaces are increasingly characterized by multimodality, nonintrusiveness, context-sensitivity, adaptivity, and teleoperability. The implementation of such properties relies on novel techniques in felds such as, e.g., computer vision, speech technology, trainable classifiers, robotics, and virtual reality.

This book puts special emphasis on technological aspects of advanced interface implementation. Furthermore it focuses on interface design and usability.

For readers with a background in engineering and computer science, most chapters offer design guidelines and case studies, as well as a description of the functioning and limitations of the algorithms required for implementation. In addition, complementary code examples in C++ are given where appropriate.

As a special feature the book is accompanied by two easy-to-handle software development environments, which offer access to extensive public domain software for computer vision, classification, and virtual reality. These environments also provide real-time access to peripheral components like, e.g., webcams or microphones, enabling hands-on experimentation and testing.

Inhalt
Non-Intrusive Acquisition of Human Action.- Sign Language Recognition.- Speech Communication and Multimodal Interfaces.- Person Recognition and Tracking.- Interacting in Virtual Reality.- Interactive and Cooperative Robot Assistants.- Assisted Man-Machine Interaction.