This paper is about implementing and analysing a strategy of generating the intermediate views of a scene from a pair of images taken from different position and orientation. In particular, the algorithm uses image mapping for pixel registration purpose and a form of interpolation to produce the in-between view. The main purpose is to note the validity and limitation of such method and to explore potential improvement and development. This work may also be a solid base for future works including the application of 3D digital video.
Farris J. Halim, Jesse S. Jin