Shinji Ogaki
@ShinjiOgaki
Followers
2K
Following
5K
Statuses
1K
A slow starter who loves simple algorithms
Japan, a dying country
Joined October 2014
@pedroth9 I'm using my hobby renderer written in c++. If you are interested in the method, I recommend you reading and its follow up papers.
0
1
6
Somehow I missed the paper “A framework for transient rendering” (. It is a very insightful paper and would be interesting to combine with methods such as EARS (.
0
11
100
I used the same trick in my nonlinear ray tracing paper🙂
@SebAaltonen Use the adjoint instead of the inverse, wouldn't that work for you? It produces the same normal direction, but not need the division, so there there are no singularity / divide by zero problems, it's faster, and can handle negative scales gracefully too.
0
6
112
@lycium No splitting at the moment. I think implementing proper energy redistribution is needed in the end..
0
0
1
@Pjbomb2 Please don't rush me! lol I'm not a DOF expert so I have to read past papers in detail. If my method has anything new, I want to write a (short) paper instead of writing every detail on X🙂
2
0
3
@Pjbomb2 @yiningkarlli @gracien_app It's a thin lens model. I just combined light sampling and lens sampling via MIS.
1
0
2
@yiningkarlli @gracien_app @Pjbomb2 No objection🙂 I even didn't think about using energy redistribution!
1
0
1
Unidirectional Path Tracing Light Sampling MIS🙂 The next step is to support more complex light paths and lenses.
1
7
79