Depth compression based on partial surface for 3D object represented by layered depth image

Abstract

Image-based modeling and rendering techniques have received much attention as a powerful alternative to the traditional geometry-based modeling and rendering techniques for generating novel views of real/synthetic scenes or objects. They use two-dimensional (2D) images as primitives to generate an arbitrary view of the three-dimensional (3D) scene or object. Since photos or images contain details of the scene, the modeling process is unnecessary. Image-based modeling and image-based rendering (IBR) techniques require proper computational resources and do not bother from the complexity of 3D objects in the scene. In addition, it is much easier to acquire a photo or a picture than complex 3D models of the scene. Among the variety of IBR techniques, a layered depth image (LDI) is one of the efficient rendering and representation methods for 3D scenes or objects with complex geometries. LDI, proposed by J. Shade et al., represents the scene or the object with an array of pixels viewed from a single reference camera location. As a data structure corresponding to a single projection of scenes or objects, it stores information for each point of intersection of each projection ray with scenes or objects. And LDI contains potentially multiple depth pixels at each discrete location in the image. Instead of a 2D array of depth pixels, a 2D array of layered depth pixels is stored in LDI. A layered depth pixel stores a set of depth pixels along one line of sight sorted in front to back order. The front element in the layered depth pixels samples the first surface seen along that line of sight; the next pixel in the layered depth pixel samples the next surface seen along that line of sight, etc. Although LDI is one of the efficient representation methods for 3D scenes or objects with complex geometries, one big obstacle is to deal with a huge amount of data, unlike ordinary 2D images. In addition, because of multiple layer characteristic of LDI, the number of layers (NOL) at...