Replies: 3 comments
|
@dunclaw, @dbuerer, @kircher1, @AdamS108, @HarelM, @oleh-derevenko, @immortalise, @jiansan, @michaeleohara, @douglascrockford, tagging a few others for awareness. |
0 replies
|
One of the grounds that our initiative must be successful is our human vision system. When we see a scene, maximally two 2D pictures from our eyes, we understand the 3D spacial structure while our brain without doing any geometrical calculation? |
0 replies
|
I saw other people discussed elsewhere at much earlier years/date than this post, would like to drop it, sorry. Will do better research next time. |
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
Hi Guys,
Do you like the idea of defining a new Spec on Pointcloud rendering?
It has been called: NeRF/Neural Radiance Fields.
As far as I understand, it's not very complex theory behind, it uses Neural Network to draw relations between 3D vertices and 2D pixels. When the training is done, we can reuse the model to generate images without Geometrical calculation.
It can solve the problems like rendering of semi-transparent surface, say, Kellyfish? And maybe for furry objects?
If you like the idea, please include me into the new board.
Cheers.
All reactions