yongbin Posted June 29, 2010 Share Posted June 29, 2010 If I have only point cloud data(not geometry), can I calculate it's occlusion in sop level? I want store them in each points. And when i render fur, i will use them. Quote Link to comment Share on other sites More sharing options...
Fabiano Berlim Posted June 29, 2010 Share Posted June 29, 2010 If I have only point cloud data(not geometry), can I calculate it's occlusion in sop level? I want store them in each points. And when i render fur, i will use them. Please. You need to be more specifc. Do you want to compute point based occlusion? Or save the the data of the raytrace occlusion into points? There's no ready to go point based occlusion in Houdini. But you can build it. (Not an easy task). Another alternative is to save the occlusion as a "irradiance cache" file. To do that you need to edit the interface of the node "mantra" and add extra parameters of irradiance. Cheers Quote Link to comment Share on other sites More sharing options...
Jason Posted June 29, 2010 Share Posted June 29, 2010 I believe there are a couple of point-based occlusion VEX SOPs buried somewhere here in the forum; although they (a) have results which are not great and ( not very fast. Quote Link to comment Share on other sites More sharing options...
yongbin Posted June 30, 2010 Author Share Posted June 30, 2010 (edited) Thank you Berlim and Jason. I am testing fur rendering. and some papers tell reusable occlusion. I tried it. Scattering fur(or pcwrite), save and pcimport that points. I calculate how many neighbors at each points. More neigbors, more darker. But the result is not good. So I want to know what (point base?) occlusion doing exactly. I didn't know that there is a way to store raytrace occlusion data into points. I saved irradiance cache to file. And when I import that file, there is many attributes in file. I don't know how to use them. And one more problem, this method makes points in camera only.(same as pcwrite) can i have all side's points? I will try both. I need compare them. Edited June 30, 2010 by yongbin Quote Link to comment Share on other sites More sharing options...
Jason Posted June 30, 2010 Share Posted June 30, 2010 NVidia hosts a point-based occlusion paper, if I remember correctly. There is a virtual disk area assigned to each point (think of a circle of radius based on the area measured by a Scatter SOP). I think there were two passes to this too, where there was a nearby set and a distant set. The occlusion is based on expose through a cone of these disks. Please search these forums too -- you'll find a couple of implementations, one in VEX and one in HDK (C++). Quote Link to comment Share on other sites More sharing options...
yongbin Posted June 30, 2010 Author Share Posted June 30, 2010 yes, i am finding and reading it now. although, it is very difficult to me. @_@ Quote Link to comment Share on other sites More sharing options...
Jason Posted July 1, 2010 Share Posted July 1, 2010 Hi there; Look in this thread: On a quick glance around I see C++ solutions shared in there; but not sure if there are any VEX versions. Quote Link to comment Share on other sites More sharing options...
yongbin Posted July 7, 2010 Author Share Posted July 7, 2010 Hi there; Look in this thread: On a quick glance around I see C++ solutions shared in there; but not sure if there are any VEX versions. Thank you! I will read this and try to understand. Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.