博客园-bssrdfhttp://www.cnblogs.com/bssrdf/zh-cnThu, 24 Jul 2008 06:09:07 GMTThu, 24 Jul 2008 06:09:07 GMT60Fluid Simulation Papers in SIGGRAPH 2008http://www.cnblogs.com/bssrdf/archive/2008/05/26/1207223.htmlBSSRDFBSSRDFSun, 25 May 2008 22:31:00 GMThttp://www.cnblogs.com/bssrdf/archive/2008/05/26/1207223.htmlhttp://www.cnblogs.com/bssrdf/comments/1207223.htmlhttp://www.cnblogs.com/bssrdf/archive/2008/05/26/1207223.html#Feedback2http://www.cnblogs.com/bssrdf/comments/commentRss/1207223.htmlhttp://www.cnblogs.com/bssrdf/services/trackbacks/1207223.html 总的来说，今年缺乏重量级的流体方面的文章。

1. Porous Flow in Particle-Based Fluid Simulations by Toon Lenaerts, Bart Adams and Philip Dutré
这篇文章考虑了porus media和流体的相互作用。以前所有的流体模拟都假设与之相互作用的物体是不可侵入的。但是实际的一些材料，比如海绵和布料，都会吸收一部分流体。这篇文章利用SPH方法， 考虑DARCY Law， 实现了流体和porus material的双向相互作用。Bart Adams去年的《Adaptively sampled particle fluids》就引起了相当大的反响。



2. Bubbles Alive by Jeong-Mo Hong, Ho-Young Lee, Jong-Chul Yoon and Chang-Hun Kim
又一篇关于气泡的文章，由一帮韩国人完成。其中的Jeong-Mo Hong以前在斯坦福跟Ron Fedkiw做，成果相当丰富。
基本思路是沿袭Greenwood and House （2004）和Lossaso et al. (2008)，用SPH驱动PLS里逃离的air particle，加入了一些其他的控制，例如vorticity confinement, 来丰富一些细节方面气泡的表现。



3. Two-way Coupling of Fluids to Rigid and Deformable Solids and Shells by Robinson-Mosher, A., Shinar, T., Gretarsson, J., Su, J. and Fedkiw, R.
这篇文章还没见到原文，但应该是值得期待的文章。流体和固体的相互作用一直是个难点，以前的一些方法尽管有效，但总是不可避免的做了很多近似和假设。不知道这次斯坦福学派有什么创新。

4. Wavelet Turbulence for Fluid Simulation by Theodore Kim, Nils Thürey, Doug James and Markus Gross
这篇比较有意思。它用小波模式把网格不能分辨的一些细微的feature加了进来，效果还不错。文中的例子都是针对烟雾的模拟，不知道能不能用于水。


5. Feedback Control of Cumuliform Cloud Formation based on Computational Fluid Dynamics
关于云的模拟，也还没有原文。

[新闻]Mozilla将于本周五发布Firefox 3.1第一个预览版]]>(Update) physically-based water simulation and animationhttp://www.cnblogs.com/bssrdf/archive/2008/05/06/1184358.htmlBSSRDFBSSRDFMon, 05 May 2008 22:48:00 GMThttp://www.cnblogs.com/bssrdf/archive/2008/05/06/1184358.htmlhttp://www.cnblogs.com/bssrdf/comments/1184358.htmlhttp://www.cnblogs.com/bssrdf/archive/2008/05/06/1184358.html#Feedback4http://www.cnblogs.com/bssrdf/comments/commentRss/1184358.htmlhttp://www.cnblogs.com/bssrdf/services/trackbacks/1184358.html  阅读全文

[新闻]瑞星将向个人用户免费1年]]>implementation of the particle levelset methodhttp://www.cnblogs.com/bssrdf/archive/2007/10/04/913787.htmlBSSRDFBSSRDFWed, 03 Oct 2007 22:07:00 GMThttp://www.cnblogs.com/bssrdf/archive/2007/10/04/913787.htmlhttp://www.cnblogs.com/bssrdf/comments/913787.htmlhttp://www.cnblogs.com/bssrdf/archive/2007/10/04/913787.html#Feedback0http://www.cnblogs.com/bssrdf/comments/commentRss/913787.htmlhttp://www.cnblogs.com/bssrdf/services/trackbacks/913787.html Just fall back from simulating water surface to correctly implementing the particle levelset (PLS) method itself. The test is the classic Zalesak's disk problem which is the rigid body rotation in a constant vorticity velocity field.


The PLS seems working correctly on this simple problem.

                                         




[新闻]中国互联网历史上最伟大的产品TOP10（二）]]>physically-based water simulation and animationhttp://www.cnblogs.com/bssrdf/archive/2007/09/11/889284.htmlBSSRDFBSSRDFTue, 11 Sep 2007 01:32:00 GMThttp://www.cnblogs.com/bssrdf/archive/2007/09/11/889284.htmlhttp://www.cnblogs.com/bssrdf/comments/889284.htmlhttp://www.cnblogs.com/bssrdf/archive/2007/09/11/889284.html#Feedback3http://www.cnblogs.com/bssrdf/comments/commentRss/889284.htmlhttp://www.cnblogs.com/bssrdf/services/trackbacks/889284.html 近几年，基于物理的流体绘制在好莱坞大行其道。其中，Particle Level Set是最具有代表性，也是最成功的一种方法。自从Nick Foster和Ron Fedkiw在2001年首次将之用于流体的模拟，各大特效公司 (ILM, PDI, PIXAR, DIGITAL DOMAIN, SONY IMAGE WORKS) 纷纷开发自己的软件版本，并用于各自的大制作电影当中。使用这种方法的大片至少有这几部：

Shrek I
Shrek II
Shrek III(?)
The Day After Tomorrow
加勒比海盗I, II
Poseidon
终结者 III
Harry Potter V

2007年的大制作就有3部：

《加勒比海盗3 》(最后的海上大战）


《蜘蛛侠3》（沙人）


《Surfs Up》（海浪）

另外，《料理鼠王》里可能也用到了。

商业软件Flowline相信也是用的Particle Level Set。

基于stable fluid solver在烟的模拟的成功，使用particle level set 方法模拟了水的运动。
网格依然很粗，40x40x40， 但大体的效果出来了。不过particle的运动好像有问题，可能是边界造成的。

不会调水的材质，用玻璃代替了。渲染依然用的pbrt。

修正了一个bug, 现在好很多了。不过粒子的运动到了边界以后还是有问题。 继续debug中........
youtube的连接

 
a pile of water falling




[新闻]盖茨官方否认天价租楼看奥运 纯属地产商炒作]]>physically-based smoke simulation and animation http://www.cnblogs.com/bssrdf/archive/2007/08/28/872970.htmlBSSRDFBSSRDFTue, 28 Aug 2007 06:28:00 GMThttp://www.cnblogs.com/bssrdf/archive/2007/08/28/872970.htmlhttp://www.cnblogs.com/bssrdf/comments/872970.htmlhttp://www.cnblogs.com/bssrdf/archive/2007/08/28/872970.html#Feedback2http://www.cnblogs.com/bssrdf/comments/commentRss/872970.htmlhttp://www.cnblogs.com/bssrdf/services/trackbacks/872970.html


这是youtube的连接，窗口大点。
http://www.youtube.com/watch?v=fbr0joBqS_o

加密了网格(100x100x100),  增大了vorticity confinement的强度，图像质量大为改观。



youtube的连接
http://www.youtube.com/watch?v=Vr0l0f4VOTM

加大烟的浓度



youtube
http://www.youtube.com/watch?v=OBUfrdW2rDo








[新闻]2008年7月24日IT博客精选]]>Bidirectional Path Tracing in PBRThttp://www.cnblogs.com/bssrdf/archive/2006/12/12/589563.htmlBSSRDFBSSRDFTue, 12 Dec 2006 01:28:00 GMThttp://www.cnblogs.com/bssrdf/archive/2006/12/12/589563.htmlhttp://www.cnblogs.com/bssrdf/comments/589563.htmlhttp://www.cnblogs.com/bssrdf/archive/2006/12/12/589563.html#Feedback9http://www.cnblogs.com/bssrdf/comments/commentRss/589563.htmlhttp://www.cnblogs.com/bssrdf/services/trackbacks/589563.htmlI had to do it by try and error. Debugging is a pain. The implementation may
still be wrong in some details, but generally it performs much better than path (PT)
tracing built-in in PBRT.

Here is the Cornell Box scene rendered with path tracing. Note the area light
pointing towards the ceiling making the scene mostly illuminated by indirect lighting.



This is the same scene rendered with my bidirectional path tracing implementation.



Both images are rendered with similar amount of time, 1400 seconds (25 samples/pixel) for bidirectional
path tracing v.s. 1510 (100 samples/pixel) seconds for pure path tracing.  However,  BPT produced much smoother
lighting than PT because it incorporates information from light sources,  just like photon mapping.

[新闻]TOM在线与Joost正式组建合资公司]]>Subsurface Scattering in PBRThttp://www.cnblogs.com/bssrdf/archive/2006/12/02/579811.htmlBSSRDFBSSRDFSat, 02 Dec 2006 06:05:00 GMThttp://www.cnblogs.com/bssrdf/archive/2006/12/02/579811.htmlhttp://www.cnblogs.com/bssrdf/comments/579811.htmlhttp://www.cnblogs.com/bssrdf/archive/2006/12/02/579811.html#Feedback6http://www.cnblogs.com/bssrdf/comments/commentRss/579811.htmlhttp://www.cnblogs.com/bssrdf/services/trackbacks/579811.html 


[新闻]财富:谷歌副总裁称其新闻搜索值1亿美元]]>Multiple Scattering of Participating Media in PBRT http://www.cnblogs.com/bssrdf/archive/2006/11/02/547433.htmlBSSRDFBSSRDFWed, 01 Nov 2006 23:41:00 GMThttp://www.cnblogs.com/bssrdf/archive/2006/11/02/547433.htmlhttp://www.cnblogs.com/bssrdf/comments/547433.htmlhttp://www.cnblogs.com/bssrdf/archive/2006/11/02/547433.html#Feedback2http://www.cnblogs.com/bssrdf/comments/commentRss/547433.htmlhttp://www.cnblogs.com/bssrdf/services/trackbacks/547433.html
This is the sponza scene with homogeneous media when only single scattering is accounted. Because there is not much direct lighting, the media is hardly to see.
 

This is the same scene as above but  when both single and multiple scattering are accounted. Now the presence of media is obvious, thanks to the light multiple scattering within the media.  This is actually a kind of global illumination for volume effect.




Update: Fixed some bugs and tested on a scene with volume caustics.

Single Scattering



With multiple scattering,  volume caustics appeared.  Small steps needed to get  volume caustics correct.

 





[新闻]Ubuntu创始人呼吁开发Linux桌面软件对抗苹果]]>Physical Sky in PBRThttp://www.cnblogs.com/bssrdf/archive/2006/06/08/420181.htmlBSSRDFBSSRDFWed, 07 Jun 2006 22:26:00 GMThttp://www.cnblogs.com/bssrdf/archive/2006/06/08/420181.htmlhttp://www.cnblogs.com/bssrdf/comments/420181.htmlhttp://www.cnblogs.com/bssrdf/archive/2006/06/08/420181.html#Feedback10http://www.cnblogs.com/bssrdf/comments/commentRss/420181.htmlhttp://www.cnblogs.com/bssrdf/services/trackbacks/420181.html
Turbidity 3.0
Time      6:00AM
Longitude 0
Latitude  20
Julian Day 120

Note: Aerial perspective is not supported now. No extra gamma correction applied.


 
.Text Admin : Editing Post "Physical Sky in PBRT"


Update: Added Sunlight as a second light source.

Gamma 1.0


Gamma 2.2


Update: Sunlight is too bright with the same scaling as the skylight. I use a larger scaling factor to get the following image

还是用建筑物来表现太阳光效果好些



[新闻]央视国际联手微软打造数字奥运媒体平台]]>First render with Maxwell 1.0http://www.cnblogs.com/bssrdf/archive/2006/04/28/387449.htmlBSSRDFBSSRDFFri, 28 Apr 2006 05:23:00 GMThttp://www.cnblogs.com/bssrdf/archive/2006/04/28/387449.htmlhttp://www.cnblogs.com/bssrdf/comments/387449.htmlhttp://www.cnblogs.com/bssrdf/archive/2006/04/28/387449.html#Feedback0http://www.cnblogs.com/bssrdf/comments/commentRss/387449.htmlhttp://www.cnblogs.com/bssrdf/services/trackbacks/387449.html

 

Render time:      28mins 40sec

Sampling level:   8.08



[新闻]Google在线百科工具Knol今日正式上线]]>