YOUR ACCOUNT

Login or Register to post new topics or replies
SpaceRay
SpaceRay

Posts: 12298
Filters: 35
Quote
ThreeDee wrote:
Frequency-modulated random tessalation. That is, tessalation patterns which are more dense at darker areas of source image.



how is the cell distribution driven?

By image luminosity, darker means more dense.



I think that this means that for example using the cells or stones component, the results could be depending on the light or dark areas of the source image like an adaptive tiling that is based on the source image, please, correct me if I am wrong

I wonder how is possible to do this in FF in some way as ThreeDee has not shown how it could be done and he is missing since long time ago

I think it would be interesting and and cool to be able to have variable sizes (density) at darker areas and less in lighter for pattern components depending on the source input dark and light areas

here below is another example of density tessellation that you can see that all is done according to the grayscale values or the light and dark.

ALSO BE ABLE TO FILL THE HOLES WITH COLORS OR SOURCE IMAGE BITS

Also would be interesting and awesome to be able to fill all these empty cells with plain colors or even better with image bits.

Thanks very much for any possible help and ideas

  Details E-Mail
Rachel Duim
So Called Tortured Artist

Posts: 2498
Filters: 188
This is doable. Already did a test with Cells, seems it should be something that can be done, I am using some of the ideas from my old Cellular Images filter. I'll post an example if I come up with something.
Math meets art meets psychedelia.
  Details E-Mail
Rachel Duim
So Called Tortured Artist

Posts: 2498
Filters: 188
This is an example using flat colored cells. The look of the 2nd example is hard to achieve, vertices don't line up along edges. This also may have been done before, but I can't remember the filter.

Math meets art meets psychedelia.
  Details E-Mail
Skybase
2D/3D Generalist

Posts: 4025
Filters: 76
Pretty sure this shouldn't be doable in FilterForge, particularly what SpaceRay's showing. Basically the scale input on any of the voronoi noise patterns are controls that can't be modified via gradient.

You can probably fake it though via masking and a few techniques just like Rachel's doing above.
  Details E-Mail
SpaceRay
SpaceRay

Posts: 12298
Filters: 35
Quote
Rachel Duim wrote:
This is doable.


Quote
Rachel Duim wrote:
This is an example using flat colored cells


Great! it looks good, although there is not much difference between darker and lighter areas, as in lighter would be bigger than in darker areas that would be smaller

will see what you with the great and high expert experience you have. Thanks really very much for taking the time to make this and appreciate it much

Quote
Rachel Duim wrote:
The look of the 2nd example is hard to achieve, vertices don't line up along edges


Of course that I do not want that you copy this example as I agree that using FF would be mostly hard to do it, but something that could be near close that could make bigger polygons on light areas and smaller on darker areas.

Quote
Skybase wrote:
Pretty sure this shouldn't be doable in FilterForge, particularly what SpaceRay's showing. Basically the scale input on any of the voronoi noise patterns are controls that can't be modified via gradient.


Read the above here, I already know is not possible to make something like this, but something near to it. I already know that the shape of the noise patterns components can't be modified easily and the example given above is made with a specific math algorithm that analyze carefully all the dark and light areas and can trace the lines in a much better way
  Details E-Mail
Rachel Duim
So Called Tortured Artist

Posts: 2498
Filters: 188
Another example. May or may not release it. Will post here if I don't.

Math meets art meets psychedelia.
  Details E-Mail
SpaceRay
SpaceRay

Posts: 12298
Filters: 35
Quote
Rachel Duim wrote:
Another example


Oh, this looks much better as there is more difference, although maybe this is not the right image to be testing this as it does not have enough difference between light and dark and would be better to use for testing on how it works a grayscale like this, so then it could be clearly seen the difference in scale of the cells that are generated

  Details E-Mail
SpaceRay
SpaceRay

Posts: 12298
Filters: 35
Or this one

  Details E-Mail
SpaceRay
SpaceRay

Posts: 12298
Filters: 35
Quote
Rachel Duim wrote:
May or may not release it. Will post here if I don't.


I think that it looks good and should be tested with more things to see really how it works and think that you may be doing a great work and would be good to put the work from an expert like you to show something that nobody else has done yet and you have been able to do it

Another example with difference between light and dark

  Details E-Mail
David Roberson
Artist
Posts: 404
Filters: 36
This reminds me of a feature I've wished for, which is scaling of noise based on map input, comparable to the way roughness is. I basically use roughness to fake it ATM, or set up a Bomber with a nice set of noise-ish particles on Lighten blending mode.

Sadly, these hacks never live up to the ideal I'm imagining...

It does make me wonder if there's an off-in-the-not-too-distant-future update that moves control inputs up to map inputs. It would be nice. Someday.

Now that I'm thinking about it, SpaceRay, doesn't Bomber/Bomber Plus allow you to place particles using a map input? Then you'd just use a spherical gradient particle, again on Lighten blending, and a source image, followed by a High Pass to get the intersection lines. Mid render ATM so I can't look.
  Details E-Mail
Rachel Duim
So Called Tortured Artist

Posts: 2498
Filters: 188
Here is a test using a left to right linear gradient. Using roughness and details in specific combinations is how I generated the five levels of cells. It's a hack, but I'll release it soon.

Math meets art meets psychedelia.
  Details E-Mail
David Roberson
Artist
Posts: 404
Filters: 36
Nice, Rachel! I look forward to seeing it!
  Details E-Mail
SpaceRay
SpaceRay

Posts: 12298
Filters: 35
Quote
Rachel Duim wrote:
Here is a test using a left to right linear gradient. Using roughness and details in specific combinations is how I generated the five levels of cells. It's a hack, but I'll release it soon.


YES!!! This is exactly was I was meaning with difference in size of the polygon shapes, and this result is really awesome and it is the first time I see this since FF 1 and nobody has been able to do it in this way that I have seen. Congratulations and well done, amazing.

it looks great and like how the empy polygons are created showing only the edges and the holes

will be waiting to see how you have done it and how it works

Thanks really very much for making this

Quote
David Roberson wrote:
Now that I'm thinking about it, SpaceRay, doesn't Bomber/Bomber Plus allow you to place particles using a map input?


Yes, Bomber allows to place particles using an grayscale or black and white map input connected to the chances, but as far as I know if you use the noise components as particles they are like image input, and you would need to modify the noise component itself before the bomber
  Details E-Mail
Rachel Duim
So Called Tortured Artist

Posts: 2498
Filters: 188
Cellular Areas

Math meets art meets psychedelia.
  Details E-Mail

Join Our Community!

Filter Forge has a thriving, vibrant, knowledgeable user community. Feel free to join us and have fun!

33,711 Registered Users
+18 new in 30 days!

153,531 Posts
+39 new in 30 days!

15,347 Topics
+72 new in year!

Create an Account

Online Users Last minute:

24 unregistered users.