Font Size: a A A

Research On High-realistic Rendering Technology For Museum

Posted on:2018-10-10Degree:MasterType:Thesis
Country:ChinaCandidate:J ZhouFull Text:PDF
GTID:2348330512999486Subject:Computer Science and Technology
Abstract/Summary:PDF Full Text Request
Digital museum represents a new existing form of cultural relics,and it is an efficient measure to carry forward national culture.At present,domestic museums have acquired a great number of digital models of cultural relics,in the future the growth of the number is going to be faster.The digital model of cultural relics usually has complicated geometric property,a huge number of patches and a high degree of detail.Because of these features of digital model,it is a difficult problem to get a high-quality rendering result of a complex digital model in a short time,and at the same time it is also an important research tropic in computer graphics.To solve the problem,this paper proposes a solution as follows:Firstly,we propose a filter based on the non-local means filter merging the pixel ray samples' histogram to denoise image by patch.This filter reduces the computation overhead during the denoising process,in order to get result image rapidly.Secondly,we propose a iterative rendering framework based on the origin Monte Carlo rendering method using SURE(Stein's Unbiased Risk Estimator)to get extra information of rendering process and guide the renderer to sample adaptively,which accelerate the rendering process greatly.Thirdly,we apply the two methods to the display system we designed,which consists of two main components,manage system and display system.Comparing with the origin Monte Carlo rendering method's result image,our system improve the e:fficiency of Monte Carlo rendering greatly and enhance the user's experience when appreciating the digital model of cultural relics.
Keywords/Search Tags:Monte Carlo rendering, cultural relics digitalizing, adaptive sampling, image denoising, non-local method
PDF Full Text Request
Related items