Efficient 3D Image Warping for Composing Novel Views

This paper presents an efficient inverse warping algorithm for generating novel views by combining multiple reference images taken from different viewpoints. The method proceeds in three steps. Firstly, the reference images are preprocessed for extracling edge pixels. Secondly, an inverse warping is performed to render the desired image from one primary reference image. By taking advantages of epipolar line features and deprh discontinuities in reference images, the inverse warping can be egiciently applied by segments, to accelerate the rendering substantially. Finally, holes in the desired image are filled up through searching the corresponding p0int.y in other reference images. At this srage, W O accelerating techniqi4e.s are presented. By rising the proposed algorithm we can navigate a virtual environment ai interactive rate.