Semiautomatic Footprints

[aaver]

For those who are interested, here is a short "tutorial" on how to make semiautomatic footprints in A:M.

Here is a short test movie using the technique.

[KenH]

Aha! Now all is clear...except the script Very educational. Thankyou. I don't know how feasible it is to make a plugin. As you say it depends on interest. As it stands, if this is the way to do footprints in AM atm, I could see a demand there to cut the time and effort. You can put my name down anyway.

[Guest jandals]

Slick! That seems way easier than trying to fake it by hand-painting in PhotoShop!

 

Rhett

[ChrisThom]

Very cool. I was also impressed with the water splash you have on your site.

 

Any insights on how that was done?

[ypoissant]

Anders,

 

This is a very interesting idea. You show a lot of enginuity. Now that you devised the main approach to this problem, I think the same idea could be performed with Photoshop with its batch processing capability and a script to merge and save the images.

[heyvern]

This is brilliant! Thanks!

 

Yves is so right! I can easily use your idea and combine it with photoshop actions and TGA sequences.

 

Think outside the box!

 

I was dieing to see how you got the footprint "image".

 

So simple you can't think of it...well I couldn't think of it. Obviously you did think of it otherwise...oh nevermind.

 

Vernon "!" Zehr

[JBarrett]

Very clever idea, and nice execution in your test! Kudos!

[aaver]

Thanks, Yves and Vern for the suggestion to use Photoshop for this effect. I had a similar thought myself, but since I only own a dated version of PSP (v.5) , in which the scripting capability is very limited I settled for Matlab instead.

 

If someone could write something on how to do this in Photoshop, I think it would be appreciated

[aaver]

Very cool. I was also impressed with the water splash you have on your site.

 

Any insights on how that was done?

I guess you are referring to this. (Thanks to Fabrice Favé for letting me use his Caustics setup.)

 

For a thorough explanation, I will probably need to write another "tutorial" and open a new thread, but the short answer is that I used the same technique as for the footprints, but with a different processing script.

 

True fluid behavior is described by the 3D Navier-Stokes equation, but this is very hard to solve and it requires massive computations. On the other hand, for most situations the full equation doesn't have to be solved. What I have tried is to solve the 2D wave equation instead. This isn't describing water at all, physically speaking, but it appears to do a reasonable good job, visually.

 

The 2D wave equation is easily solved by a number of standard methods. I have chosen to implement a finite difference solver, which took me less than an hour to derive and implement.

 

If the programmers in this forum want something to toy with, they might try out the following iteration scheme:

 

   // u0 and u0' are the initial values for the position and velocity respectively
   // k is a material dependent parameter

   U[1] = u0;
   U[2] = u0 + k*u0';
   for (int n = 2; n<N; n++) {
       for (int i = 1; i<Ni-1; i++) {
           for (int j = 1; j<Nj-1; j++) {
               U[n][i][j] = U[n-1][i][j] - U[n-2][i][j] + 0.25*(U[n-1][i+1][j] + U[n-1][i-1][j] + U[n-1][i][j+1] + U[n-1][i][j-1]);
           }
       }
       saveFrame(U[n]);
   }

 

For calm water, u0 and u0' will be zero all over. Without further explanations you could feel safe to set k=1.

 

Needless to say, I'm only experimenting with this so far so take it for what it is.

[SeanC.]

::Stares blankly ahead smiling, clearly with no idea of what you just said::

[aaver]

::Stares blankly ahead smiling, clearly with no idea of what you just said::

Sorry about that I realize I'm not doing a good job explaining this.

However, if you accept that there is a standard method available which solves something called the "2D wave equation", what I am saying is that this "2D wave equation" approximates a dynamic water surface, visually - meaning that the result looks somewhat like water, which is all we want in this community.

[KenH]

Bump

 

Aaver: Did you ever get round to developing a plugin for doing footprints in the sand? It would come in handy right about now. Thanks.

[aaver]

Bump

 

Aaver: Did you ever get round to developing a plugin for doing footprints in the sand? It would come in handy right about now. Thanks.

Sorry, but no, not yet.

I haven't forgot about it though

 

Depending on what you need, in the meantime maybe you could try to use Photoshop as suggested by Yves earlier in this thread?

[Zaryin]

First time I've seen this thread. Very ingenious.