Multi-Viewpoint Videos Merging System using Auto-Stereoscopic Display in Tele-Immersion

H. Hazama, N. Sakamoto, H. Horii, Y. Ebara, and K. Koyamada (Japan)

Keywords

Teleimmersion, Facetoface communication, Autostereoscopic 3D display, Eyetoeye contact, Merging of multiviewpoint image,

Abstract

In order to achieve reality and visual depth during group to group communication, interactive communicati on in real time and 3-D (3-dimensional) display are indispensable. Generally, 3-D display requires deflection glasses and a head mount display (HMD) [1]. In such a condition, it is usually difficult to carry out face-to-face communication because people cannot establish eye-to eye contact with others. In this paper, we have designed a system that can merge multi-viewpoint video images in real time. The merged video image can be displayed in 3 D by using the auto-stereoscopic display system [2]. Next, we have examined how well the students can establish eye-to-eye contact with the person on the 3-D display during a face-to-face communication. From this result, we have showed that our new system can contribute to enhance reality of face-to-face communication. Moreover, we propose a new system which can merge visualization images obtained from multi-viewpoints so that one visualization image can also be displayed. In addition, we have implemented the function that can merge an already merged video image with an already merged visualization image in real time.

Important Links:



Go Back