基于AdaBoost.MH的Reyes渲染架构时间预估算法

计算机科学 ›› 2014, Vol. 41 ›› Issue (2): 76-81.

基于AdaBoost.MH的Reyes渲染架构时间预估算法

孟庆利,吕琳,靳颖,孟祥旭,孟雷

山东大学计算机科学与技术学院济南250101;山东大学计算机科学与技术学院济南250101;山东大学计算机科学与技术学院济南250101;山东大学计算机科学与技术学院济南250101;南洋理工大学计算机工程学院新加坡

出版日期:2018-11-14 发布日期:2018-11-14
基金资助:
本文受863重大项目子课题(2012AA01A306),国家自然科学基金(61202147),山东省自然科学基金(ZR2012FQ026)资助

Time Prediction for Reyes Rendering Architecture Based on AdaBoost.MH Algorithm

MENG Qing-li,LV Lin,JIN Ying,MENG Xiang-xu and MENG Lei

Online:2018-11-14 Published:2018-11-14

摘要/Abstract

摘要： 在大规模真实感渲染系统中,需要对渲染任务进行分解和调度,将其优化后分配给不同的可用计算资源,实现快速集群渲染。为了实现渲染任务的有效分解和调度,提高并行效率,高精度的时间预估算法是不可欠缺的。通过深入研究使用RenderMan规范的渲染器常用的Reyes渲染架构中对渲染时间产生影响的各种因素,分析提取出影响渲染时间的7大要素特征,提出了基于AdaBoost.MH的渲染时间预估算法。通过在基于Reyes渲染架构的渲染引擎中的实验与测试表明,训练集和测试集的准确率分别达到79%和78%,为渲染任务的并行调度奠定了基础,同时也为渲染费用预估提供了依据。

关键词: 时间预估,AdaBoost.MH算法,Reyes渲染架构,集群渲染中图法分类号TP391文献标识码A

Abstract: A high performance computer system,e.g．a computer cluster,built for large-scale photorealistic rendering,i.e．render farm,is a basic infrastructure for producing CG animations and movie special effects．In a render farm,one of the key issues is the strategy of scheduling and dispatching rendering jobs,which greatly affects the computing efficiency．Time prediction for a render job plays an important and essential role in the job scheduling and dispatching stage．However,there is no feasible algorithm and even little research work on this problem．We focused on the Reyes rendering architecture．We first analyzed the factedors that affect the rendering time and extracted the seven key features as the feature vector based on the analysis．Then we proposed a time prediction framework based on AdaBoost.MH algorithm,in which we transformed the rendering time into intervals and combined them with the feature vector to obtain the samples．Experimental results show the effectiveness of the algorithm,and the accuracy of training set and test set is 79% and 78%．

Key words: Time prediction,AdaBoost.MH algorithm,Reyes rendering architecture,Render farm

孟庆利,吕琳,靳颖,孟祥旭,孟雷. 基于AdaBoost.MH的Reyes渲染架构时间预估算法[J]. 计算机科学, 2014, 41(2): 76-81. https://doi.org/

MENG Qing-li,LV Lin,JIN Ying,MENG Xiang-xu and MENG Lei. Time Prediction for Reyes Rendering Architecture Based on AdaBoost.MH Algorithm[J]. Computer Science, 2014, 41(2): 76-81. https://doi.org/

参考文献

[1] Reinhard E,Jansen F W．Scheduling Issues in Parallel Rendering[C]∥Proceedings of the First Annual Conference of the Advanced School for Computing and Imaging,May 1995:268-277
[2] Chong A,Sourin A,Levinski K,et al．Grid-based computer animation rendering[C]∥Proceedings of the 4th international conference on Computer graphics and interactive techniques in Australasia and Southeast Asia,2006．Kuala Lumpur,Malaysia:ACM 2006:39-47
[3] Cook R,Carpenter L,and E．Catmull．The Reyes Image Rendering Architecture[J]．ACM SIGGRAPH Computer Graphics,ACM,1987,21(4):95-102
[4] Apodaca A A,Gritz L．Advanced RenderMan creating CGI for motion pictures[M]．Francisco:Morgan Kaufmann Publishers,2000:137-137
[5] Litke A,Tserpes K,Varvarigou T．Computational workload prediction for Grid oriented industrial applications[C]∥ IEEE Internationale Symposium on The case of 3D-image rendering:Cluster Computing and the Grid,2005,CCGrid2005．IEEE,2005,2:962-969
[6] Doulamis N,Doulamis A,Panagakis A,et al．A combined fuzzy-neural network model for non-linear prediction of 3-D rendering workload in Grid computing[J].IEEE Transactions on Systems,Man,and Cybernetics,Part B:Cybernetics,2004,34(2):1235-1247
[7] Doulamis N,Doulamis A,Panagakis A,et al．Workload Prediction of Rendering Algorithms in GRID Computing[C]∥European Multigrid Conference．Hohenwart,Germany,2002:7-10
[8] 张自敏,樊艳英,陈冠萍.改进的BP神经网络在地方GDP预测中的应用[J].计算机科学,2012,9(Z11):108-110,7
[9] Pixar．The Renderman Interface[S]．Version 3.2.1specification,San Rafael,California:Pixar,November,2005:4-48
[10] Mitchell T M．机器学习[M]．曾华军,张银奎,等译．北京:机械工业出版社,2003:1-10
[11] Freund Y,Schapire R E．Adecision-theoretic generalization ofon-line learning and an application to boosting[C]∥Computational Learning Theory．Berlin:Springer-Verlag,1995:23-37
[12] Freund Y,Schapire R E．Experiments with a new boosting algorithm[C]∥Machine Learning-International Workshop then Conference．San Francisco:Morgan Kauffman,1996:148-156
[13] Schapire R E．A brief introduction to Boosting[C]∥InternationalJoint Conference on Artificial Intelligence．San Franscisco:Morgan Kaufmann Publishers Inc,1999:1401-1406
[14] Wikipedia．AdaBoost．http://en.wikipedia.org/wiki/AdaBoost,2013-04-15
[15] Schapire R E,Singer Y．Improved Boosting Algorithms UsingConfidence-rated Predictions[J]．Machine Learning,1999,37(3):297-336
[16] JFriedman J,Hastie T,Tibshirani R．Additive logistic regres-sion:a statistical view of boosting[J]．Annals of Statistics,2000,28:337-407
[17] Cortes R,Raghavachary S．The RenderMan Shading LanguageGuide[M]．Thomson Course Technology,2007:237-241
[18] Benbouzid D,Busa-Fekete R,Casagrande N,et al．MultiBoost:a multi-purpose boosting package[J]．Journal of Machine Lear-ning Research,2012,13:549-553

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed