授与機関名広島大学(Hiroshima University)
一般注記type:text
Three dimensional video ( 3DV ) and multi-view imaging technologies may be the next step in the evolution of motion picture formats, as we presently witness the appearance of 3D displays, multicamera systems with dense or sparse camera configuration, coding systems. Going with the demand of entertainment and progressive development of digital devices, developing 3D processing algorithms, related applications and systems have been attracted extensive attentions in the industrial and research communities. Depth inference from stereo and multi-view images is one of the most fundamental techniques in 3D digital imaging applications since it provides the perception and visualization of the real word environment in 3DV as well as a useful cue for other applications. This thesis devotes to firstly study depth estimation from multi-view images and then use this useful information for three applications including: one of the key applications in 3DTV namely free viewpoint synthesis, and other two applications object segmentation and multiple moving object tracking.……
連携機関・データベース国立情報学研究所 : 学術機関リポジトリデータベース（IRDB）（機関リポジトリ）