View Generation for 3-d Scenes from Video Sequences

This paper focuses on a representation for 3-D scenes consisting of dense depth maps at preselected viewpoints from video sequences derived from unknown but approximately horizontal motion. In contrast to existing methods that construct a full 3-D model or those that exploit geometric invariants, an intensity-depth representation is used to generate arbitrary views of the 3-D scene. Speciically, the depth maps are regarded as vertices of a deformable 2-D mesh which are transformed in 3-D, projected to 2-D, and rendered to generate the desired view. Experimental results are presented to verify our approach.