Automatic multi-camera video composition

作者: Li-Cheng Tai , Boon Lock Yeo , Minerva Yeung

DOI:

关键词:

摘要: Methods and apparatus are disclosed for automatically selecting a video output (50) from among several input sources (C1-C3). In one method, or more audio sensors (A1-A3) associated with each source. Preferably, an sensor is positioned to receive signals directions that favorable coverage in the field of view An autoselector (40) calculates scores over short (e.g., 0.5 seconds) examination intervals. At interval, potential exists different source be selected as output. The selects based on well recent time-history selection. For instance, if new has just been selected, selection may disabled few seconds. also used increase probabability varies seemingly-natural manner.

参考文章(7)
Peter L. Chu, Steven L. Potts, Hong Wang, Wendi Beth Rabiner, Locating an audio source ,(1999)
Takehiko Yamada, Masaaki Takizawa, Kiyoshi Ishida, Atsuo Yoshida, Yoji Shibata, Atsushi Ishibashi, Hitoshi Matsushima, Teleconference terminal equipment and teleconference module ,(1992)
John J. Hildin, Voice-following video system ,(1997)