bssrdf

@LoveGraphics
我直接ray trace的level set, 所以这一步就省略了。

@yuanhong
pbrt可以渲染动画的，批处理就是了。

请问，你对target density进行的使用高斯kernel的convolution是怎么实现的？
你使用的参数的值分别是多少(sigma, vf, vd 和 vg)？
谢谢

@yuanhong
point repulsion是Greg Turk在他的博士论文"Texturing Surfaces Using Reaction-Diffusion"里提出的方法。 这种方法可以在一个多边形模型上均匀取点。这样取的点就可以满足Jensen方法的要求。可惜我的point repulsion实现还是有些问题。现在我必须取比[A/(pi*lu*lu)]更多的点。
至于计算irradiance， 我PHOTONMAP和DIRECTLIGHTING都用了。效果差不多。而PHOTONMAP我只用了DIRECT MAP。我想interreflection部分并不重要。

你可以在google talk找到我：bssrdf@gmail.com

@yuanhong
你实现的效果也很好呀！
我用的就是Jensen and Buhler（2002）那篇文章。很多人的实现都是基于这片文章，可能是因为比较容易变成代码吧。buddha scene在我的机器上跑的时间不长，具体的时间不记得了。最费时间的是预处理。
一个问题，你在你的实现里是怎么取的sample point？你用了point repulsion了吗？还有，你计算irradiance的时候，用的是photon map还是别的方法？
很高兴和你交流一些心得体会。

@mz
I am doing it out of interest and yes, I read a lot of papers.
PBRT book is worthwhile to get. Without it, it's going to be a nightmare to implement these more advanced algorithms. Of course you can use many open source ray tracers, but none of them has a detailed explanation as in PBRT.

@MLJack
The glass is just the combination of specular reflection and transmission as defined in PBRT.
As Veach's thesis shows, the weight of the path when there are specular vertices is easy to calculate by setting corresponding PDFs to zero. However, the problem with these specular vertices is that no connection made to them such that some effects associated with them can only be obtained by standard path tracing. For those parts of the image, BPT is as noisy as PT, e.g., the glass sphere.

Haven't tried glossy BTDF yet. Interesting to see though.

@AdamGamer
No way for this kind of images to be rendered in real time, at least not in the near future.

Well, the most useful reference I found is Jensen's book "Realistic image synthesis using photon mapping". Jensen and Christensen's 1998 paper is also helpful. Another worthwhile book is "Advanced global illumination".

@billconan
Maxwell is a Monte Carlo ray tracer using MLT algorithm. Pbrt is also a MC ray tracer, only that it uses other global illumination algorithms, e.g., Path Tracing, Irradiance Caching, Photon Mapping. You can make Pbrt implementing MLT, but a lot of work has to be done.

@li kong
I wrote a function which transforms SPD (spectrum data) to RGB through XYZ.

First you use CIE X,Y,Z data to convert SPD, which is a series of number corresponding to their wavelength, to XYZ. Then XYZ to RGB is straightforward.

@li kong
The methods are described in the paper I mentioned above in very much details.

Google that paper's name will give you some additional information, especially a post on gamedev.net.

The paper is "A Practical Analytic Model for Daylight" by A. J. Preetham et al. (1999). The method from this paper has been used in Nextlimit Maxwell, the most advanced ray tracer today.

All the equations are listed in the paper, along with the c++ source code on their website (http://www.cs.utah.edu/vissim/papers/sunsky/).

@mljack_
I am using PBRT (www.pbrt.org) code to implement the MLT algorithm. There is still much work to be done.

申请加入