立體渲染(Volume Rendering)在醫學方法相當普遍，而直接渲染立體影像在過去曾經是一件計算量相當龐大的一件工作，所以在過去研究直接立體渲染相關的研究並不多。但是，由於現今電腦硬體以及演算法的發展迅速，立體渲染已經變成一個相當普遍的一種渲染方法，即使在個人電腦上也能夠有一定的繪圖效率，再加上nVidia公司所提出的CUDA語言，也就是程式設計師可以很簡單的利用繪圖處理器來撰寫平行處理程式，使得立體渲染的研究在近年來是愈來愈多，執行的速度也因此愈來愈迅速。
本篇論文主要是利用資訊理論的方式來尋找一個立體影像的最佳視點，首先先找出兩個全域最佳視點，接著利用K-Mean Clustering方法來分出一個立體影像中重要性較高的部分，再利用Mean Shift Clustering自動分出數個重要性較高的群集點，而這些群集點就有可能是包含細部變化量大的區域，也就是我們希望看到的重要的細節，最後透過找出這些重要區域的最佳視點後，把所有找出的最佳視點利用傳統排版風格的編排來將圖片呈現在使用者的面前。

Volume Rendering is a general purpose in the medical field. In the past, directly volume rendering was a huge computation job, it was actually few that studied and played up relevant research. Recently, the hardware and the algorithm are high development, volume rendering become the popular method in medical field. Even in the personal computer can also have a certain drawing efficiency. By the CUDA language which is proposed by nVidia Corporation, programmers can easily use GPU to write the parallel processing program. Therefore, research on volume visualization has been more popular in recent years, and the execution speed is faster and faster.
This paper uses the information theory to find the multiple viewpoint of a volume data. First of all, we find the two global best viewpoints. Then we will get the importance threshold from a volume data by the K-Mean Clustering method; using the importance threshold and the Mean Shift Clustering method to get some high important clusters automatically. Clusters we found may be changed wildly in the detail region, and that is what we are interested in. We can find the best viewpoint from the inside of these clusters. Finally, we manage to find all of best view and present to viewer according to traditional layout style.

T. Arbel and F. Ferrie. Viewpoint selection by navigation through entropy maps. In Proc. of the 7th IEEE International Conf. on Computer Vision (ICCV-99), volume I, pages 248–254. IEEE, 1999

S. Takahashi, I. Fujishiro, Y. Takeshima, and Tomoyuki Nishita. A feature-driven approach to locating optimal viewpoints for volume visualization. In Proc. of IEEE Visualization ’05, page TBA, 2005.

P. P. V’ azquez, M. Feixas, M. Sbert, and W. Heidrich. Viewpoint selection using viewpoint entropy. In Proc. of Vision, Modelling, and Visualization ’01, pages 273–280, 2001.

P.-P. V’ azquez, M. Feixas, M. Sbert, andW. Heidrich. Automatic view selection using viewpoint entropy and its application to image-based rendering. Computer Graphics Forum, pages 689–700, 2003.

Guangfeng Ji and Han-Wei Shen. Dynamic View Selection for Time-Varying Volumes. In IEEE Transactions on Visualization and Computer Graphics, Vol. 12, No. 5, 2006

U. D. Bordoloi and H.-W. Shen. View selection for volume rendering. In IEEE Visualization Conference 2005, pages 487–494, 2005.

V. Blanz, M. J. Tarr, and H. H. B‥ ulthoff. What object attributes determine canonical views. Perception, 28(5):575–600, 1999.

Kwan-Liu Ma. Image Graphs – A Novel Approach to Visual Data Exploration. IEEE Visualization Proceedings of the conference on Visualization ’99. 81-88 1999

Daniel Lewis, Steve Haroz, and Kwan-Liu Ma. Layout of Multiple Views for Volume Visualization: A User Study. ISVC 2006, LNCS 4292, pp. 215-226, 2006

S. Abbasi and F. Mokhtarian. Automatic view selection in multi-view object recognition. In Proc. of Int. Conference on Pattern Recognition, volume 1, pages 13–16, 2000.

S. Fleishman, D. Cohen-Or, and D. Lischinski. Automatic camera placement for image-based modeling. Computer Graphics Forum, 19(2):101–110, 2000.

G. Kindlmann and J. W. Durkin. Semi-automatic generation of transfer functions for direct volume rendering. In Proc. of IEEE Symposium on Volume Visualization ’98, pages 79–86, 1998.

G. H. Weber and G. Scheuermann. Automating transfer function design based on topology analysis. In Geometric Modeling for Scientific Visualization, pages 293–306. Springer-Verlag, 2004.

L. Wong, C. Dumont, and M. Abidi. Next best view system in a 3-d modeling task. In Proc. of International Symposium on Computational Intelligence in Robotics and Automation (CIRA), pages 306–311, 1999.

N.A. Massios and R.B. Fisher. A best next view selection algorithm incorporating a quality criterion. In Proc.of the Britsh Machine Vision Conference, 1998.

D.R. Roberts and A.D. Marshall. Viewpoint selection for complete surface coverage of three dimensional objects. In Proc.of the Britsh Machine Vision Conference, 1998.

V. Hlavac, A. Leonardis, and T.Werner. Automatic selection of reference views for image-based scene representations. In Lecture Notes in Computer Science, pages 526–535, New York, NY, 1996. Springer Verlag. Proc. of European Conference on Computer Vision ’96 (ECCV ’96).

P. Barral, G. Dorme, and D. Plemenos. Scene understanding techniques using a virtual camera. In A. de Sousa and J.C. Torres, editors, Computer Graphics Forum (Proc. Eurographics ’00), volume 19, 2000.

P.-P. Vaquez, M. Feixas, M. Sbert, and A. Llobet. Realtime automatic selection of good molecular views. Computers and Graphics, 30:98–110, 2006.

Stefan Bruckner and M. Eduard Groller. Exploded Views for Volume Data. In IEEE Transactions on Visualization and Computer Graphics, Vol. 12, No. 5, 2006.

I. Viola, A. Kanitsar, and M. E. Groller. Importance-driven feature enhancement in volume visualization. IEEE Transactions on Visualization and Computer Graphics, 11(4):408-418, 2005.

I. Fujishiro, Y. Maeda, H. Sato, and Y. Takeshima. Volumetric data exploration using interval volume. IEEE Trans. on Visualization and Computer Graphics, 2(2):144–155, 1996.