finally got it working in openCL:
Attachments in this post:
http://fx-td.com/houdiniandchill/wp-content/uploads/discord/20241412/28/24/image.png
http://fx-td.com/houdiniandchill/wp-content/uploads/discord/20241412/28/24/CAMERA_PROJECTION_COPS_OCL.hipnc
just brute force grabbing eaach matrix component
not sure why i need to do the minus sign in Y tho… guess some kind of handedness oddity
This one updates much more consistently compared to the vex one and feels really fast.
oops. the UV space was a little wonky there…
add this line
Attachments in this post:
http://fx-td.com/houdiniandchill/wp-content/uploads/discord/20244812/28/24/image.png
here is a simple vex camera projection if you do feel the need to be overcomplicated. 🙂
this is useful if your pos data comes from a rendering or something.
Attachments in this post:
http://fx-td.com/houdiniandchill/wp-content/uploads/discord/20241612/27/24/houdini_0NwyzI106U.mp4
http://fx-td.com/houdiniandchill/wp-content/uploads/discord/20241612/27/24/CAMERA_PROJECTION_COPS.hipnc
i like filmot.com for doing something similar, it’s a search engine for youtube subtitles mainly but the builtin player shows the subs as a nice little column you can scroll through really fast and click the little button at the start of each line to jump in the video to that point
it’s kind of insane that you can search for like “vellum configure strut” then keep hitting “next clip” and it’ll jump through every single time someone has talked about vellum struts on all of youtube
Attachments in this post:
http://fx-td.com/houdiniandchill/wp-content/uploads/discord/20240612/24/24/Screenshot_2024-12-24_at_13.27.55.png
@tokyomegaplex ?
a bit of this and that :
Tested with that FBX that has embedded textures … at some point as folder “`Gas Station.fbm“` was created but I don’t by who, probably the fbx_archive_import … after that you can use a “attribfromparm” that helps to create a dict from nodes parameters, and retrieve the path to every texture plug as basecolor_texture.
It’s possible you don’t even need a loop for this, but it’s easier to build and check strings that way.
Attachments in this post:
http://fx-td.com/houdiniandchill/wp-content/uploads/discord/20240212/23/24/image.png
http://fx-td.com/houdiniandchill/wp-content/uploads/discord/20240212/23/24/FBX_Texture_To_Cd.hiplc
http://fx-td.com/houdiniandchill/wp-content/uploads/discord/20240212/23/24/Gas_Station.FBX
if you want to sample points at the medial axis you could look at something like `coverage axis ++` which uses some really simple metrics to sample points at the medial axis without needing to do any kind of fusing/remeshing or advection.
But i will say the medial axis is inherently not as useful as the straight skeleton, cuz in 3d for many shapes its a surface and not a curve…
arxiv.org/abs/2401.12946 its a neat paper, a totally different approach for finding these surfaces. @mattiasmalmer @_white_dog_
if you want to get topology out of this and not just a point cloud it definitely takes more effort, and no one likes using effort
Attachments in this post:
http://fx-td.com/houdiniandchill/wp-content/uploads/discord/20243212/22/24/image.png
http://fx-td.com/houdiniandchill/wp-content/uploads/discord/20243212/22/24/coverage_axis_plus_plus_JR.hiplc
if you want to sample points at the medial axis you could look at something like `coverage axis ++` which uses some really simple metrics to sample points at the medial axis without needing to do any kind of fusing/remeshing or advection.
But i will say the medial axis is inherently not as useful as the straight skeleton, cuz in 3d for many shapes its a surface and not a curve…
arxiv.org/abs/2401.12946 its a neat paper, a totally different approach for finding these surfaces. @mattiasmalmer @_white_dog_
if you want to get topology out of this and not just a point cloud it definitely takes more effort, and no one likes using effort
Attachments in this post:
http://fx-td.com/houdiniandchill/wp-content/uploads/discord/20244312/22/24/image.png
http://fx-td.com/houdiniandchill/wp-content/uploads/discord/20244312/22/24/coverage_axis_plus_plus_JR.hiplc
Hehe I’ll get right on that
here’s the hip. it’s pretty basic and not even technically a solver. It’s a surfacedist lookup from all points to a source group of points (the black rectangle in the beginning). Then I’m animating the visualization range of the distance value. Increase the grid density / decrease the mountain element size for more tendrils. If there’s a way to speedup that surface calculation, this should run even faster.
Attachments in this post:
http://fx-td.com/houdiniandchill/wp-content/uploads/discord/20245412/22/24/overview.jpg
http://fx-td.com/houdiniandchill/wp-content/uploads/discord/20245412/22/24/noise_infection_solver-basic.hipnc
Way too organic, I think the first level should be done using one of this findshortestpath to create almost tree branch pattern ?
Attachments in this post:
http://fx-td.com/houdiniandchill/wp-content/uploads/discord/20245812/21/24/image.png
http://fx-td.com/houdiniandchill/wp-content/uploads/discord/20245812/21/24/pattern.hiplc
