Research On Surface Simplification Algorithm And Progressive Transmission Method Of 3D Model

Large-scale 3D (three-dimensional) scene rendering and transmission are important techniques in the 3D internet development. The data of 3D mesh models is too much and network bandwidth is limited, which restrict the development of large-scale 3D scene rendering and transmission.3D mesh simplification technology is used to reduce the number of 3D mesh data while maintaining the original model features. Using this technique can greatly reduce the data amount when rendering three-dimensional scene. Progressive transmission technology means that the details of 3d model are encoded in a data stream, which allows a progressive refinement of the model from an approximation to the original high quality model. Using this technique, the client side can render without waiting and the bandwidth requirements can be significantly reduced. The two technologies were studied and the effective integration of these two technologies was shown in this thesis.In the aspect of multi-resolution three-dimensional mesh model generation, this thesis introduced a variety of generation ideas, we chose the most widely used edge collapse method as a way of basic simplification operations. Then a simplification algorithm based on edge length and curvature and QEM (quadratic error measure) simplification algorithm were analyzed. During the study of algorithmic processes, firstly, this thesis implemented two algorithms. Secondly, by using Hausdorff distance and visual analysis, the advantages and disadvantages of these two categories were analyzed. Finally, on the basis of these two types of algorithms, we used edge length and surface area to optimize the algorithm based on half-edge QEM.In the aspect of progressive transmission, the principle of progressive transmission was introduced, the information required for progressive transmission was analyzed and the most important collapse recording format was designed and implemented. A system of C/S model was designed in this paper to verify the correctness of the collapse recording format. The results showed that the collapse recording format met the need of progressive transfer of dimensional mesh models in a variety of scenarios. In addition, in order to facilitate applying this technique to the actual project, the LRU (Least Recently Used) cache was used to improve server responsiveness, Huffman coding was used to compress data and multithreading mode was used to provide multi-client concurrent link.