Publication Details

International Conference on Computer Games, Animation, and Multimedia, CGAT 2008, Singapore

Contribution To Book Anthology


In this paper we overview a novel mesh representation method, entitled MESHGRID. MESHGRID is a compact, multi-scalable and animation-friendly representation system of 3D objects, standardized in MPEG4 AFX. One of the particularities that distinguish MESHGRID and broaden its capabilities compared to the existing techniques is the very manner in which a 3D object is represented. MESHGRID is able to capture, store and reconstruct not only the surface of the 3D object to be encoded, but also information about its volume. Therefore MESHGRID allows for two types of scalability in both view-dependent and view-independent scenarios, including: (1) resolution scalability, i.e. progressive level of details of the wireframe, and (2) shape precision, i.e. progressive precision of the volumetric description. Additionally, MESHGRID is capable of representing very efficiently animations and dynamic meshes, based on techniques involving the alteration of the volume, besides the classical vertex-based methods. Furthermore, MESHGRID's bitstream is suitable for very efficient data transmissions over varying-bandwidth channels in client/server application scenarios. That is, only the desired regions of interest of the 3D object and the necessary levels of detail can be requested, implying that only the needed parts of the MESHGRID's bitstream are being transmitted. One concludes that MESHGRID is suitable for a large variety of applications in the field of 3D graphics, especially where the polyvalent scalability and the efficient transmission of the 3D objects are essential requirements.