SUMMARYThe authors developed a new vision-based interface that detects gestures in real time to enable interaction between the user and computer. This vision-based interface consists of two elements: the "Motion Processor TM " and the "extraction of the region of interest (ROI)." The Motion Processor is a new image acquisition device. By illuminating the target object with near-infrared light and using image sensors to collect the reflected light, the Motion Processor can eliminate the background and collect shape, motion, and depth information of only the target object. In addition, the authors proposed a method of using the collected image depth information to quickly and stably extract the ROI and showed, according to evaluation experiments on a PC, that the region can be detected with high precision within 0.06 second. Real-time sensing of the shape and motion of a specific object from an image can be implemented by using the Motion Processor and this ROI extraction method, enabling this interface to function as a computer eye that can easily recognize a target object. The authors applied this real-time vision-based interface to the field of robotics and used it together with speech recognition on a single PC to create a prototype of a pet robot system that can respond to the user's gestures.

We developed a new vision-based interface that can detect the human motion and control a pet robot interaction system. This vision-based interface is composed of the "Motion Processor (MP)" and the "Region Of Interest (ROI)". The MP is a new image input device. The MP can exclude background part and capture only the object's shape, motion and depth information in real time. The ROI is a concept in order to distinguish objects in a captured image. For detecting the ROI, the proposed method used depth information of the captured image and offered a fast and robust searching method. It enables the MP to sense a variety of object's shapes, motions, positions and so on. Therefore, our vision-based interface using the MP with the ROI can easily recognize object's motions as effective computer eyes. The user can interact with the pet robot (constructed by LEGO MINDSTORMS TM) using this vision-based interface.

People naturally use gesture to aid inter-personal communication. However, during computer interaction, users are obliged to use conventional devices such as keyboards, mice, and game-pads. In this paper a motional interface is described that allows processing of gestural inputs through the medium ofa 3D image-input device, or Motion Processor, which provides a more natural framework for human-computer interaction. Several PC applications for this real-time motionai interface will be presented in the live demonstration session. These applications are edutainment prototypes that make use of natural hand movement.

We have developed a new content-finding navigation system "Roaming Navi", in which subjective relevance between TV contents is accurately predicted by a complex non-linear formula. In order to optimize the formula, it was necessary, but difficult, to obtain enormous reliable subjective data on the relevance between TV contents. To tackle this problem, we proposed a new probabilistic method to handle subjective data, and the validity of the method was confirmed in empirical questionnaire data. Then, we conducted optimization of the prediction formula combining various optimization methods. As a result, the accuracy of the prediction was significantly improved. Moreover, these methods are broadly applicable in the discipline of Kansei engineering.

No abstract