Pont du Gard
-
Fantastic!
By the way Julius, you are from Italy. I am from Spain.
What the... hell are we doing modelling in France?
-
@pichuneke said:
Fantastic!
By the way Julius, you are from Italy. I am from Spain.
What the... hell are we doing modelling in France?
ehe...
-
How did you make the last one? It's impressive, I want to learn!
-
Those are AWESOME renders..!! 
-
This is GREAT Julius!
-
i'm glad you like them.
pichuneke, here's how i did the last one:
1- choose the photo where you want to 'put" your model
2- match the model with the photo with the photomatch function in SU. adjust the shadows so that they match as well as you can with those of the photo (to this purpose it would be better to have already placed the model in Goggle Earth. also it's helpful if you know the day and the time when the photo was taken. in my case, i got the photo from the web, so i had to proceed without this help.)
3- export the model to your render application. in this case i use Kerkythea. i used this export settings:
geometry: yes
lights: yes
clay model: no
photomatched: no
default UVs: no
4- render the scene corresponding to the matched photo with the default resolution (955x613)
5- when you're ready with the render, go back to SU. turn off the photo on the background. go to the scene with the matched photo.
turn off shadows and ope the "styles" window. among the "default styles" choose "02 HiddenLine". then edit the color of the edges (in HLS mode) and adjust the luminosty to about 97. then open the "Fog" window: uncheck "use background color" and turn the color from white to black simply setting the luminosity to 0. use the two slider to get the background to be black and the part of the model you want to put in the photo, white. eventually export a 2d graphic with the default resolution (955x613).
if you didn't understand this step, take a look at this video (just the first part, though also the second part may help):[flash=425,355:3s79oftd]http://www.youtube.com/v/fDzNJYi6Bok[/flash:3s79oftd]
6- before going to PhotoShop, resize the photo: keeping the original ratio, adjust ONLY the height to 613.
7- open in PS the three images: the render, the "foggy model", the real photo.
(notice the resoultion)
now select the render: in the "layers" palette double click the only layer existing ("Background") and a window will appear. make sure to have the default settings (name: layer 0; color: none; opacity: 100%) and click ok. select "add layer mask" at the bottom of the "layers" palette (third icon from the left). holding the "alt" key, click on the layer mask you've just created.
8- select the foggy image: go to "select -> all" then "edit -> copy" and eventually go back to the render "edit -> paste".
9- in the layers palette click on the colored image and you will see that the parts of the image corresponding to the black color in the layer mask are transparent. go to the "channels" palette and select the "layer 0 mask" channel. go to "select -> color range" and set fuzziness to 200. select the "RGB" channel and then using the move tool drag the selected part of the image to the real photo.
10- now select the clone stamp tool. select the "background" layer and (holding the alt key)
click in a certain point on the image. now go to "layer o", click in the same point and then, holding the mouse left button, drag the brush. for example if you have a tree in front of an house, you will have to drag the brush over that tree (which is hidden by the house. to this purpose it's helpful to adjust the opacity of "layer 0").
11- when you're done save the image both as a .psd (in case you will need to edit it) and as a .jpg.hope you understand, even though it's likely i did many grammar mistakes
-
THANKS!
I am going to save this page, to do it in the future.
-
Julius,
This is a wonderful TUTORIAL! Thanks for sharing these steps!
(BTW did you know that you can even "clip render" your model in KT?)Would you mind copying it (with the original image) into the Tutorial Forum?
-
-
It's Fletch:
Video (Flash) - PDF - Forum post
-
cool...though i think for the tutorial it's better the "depth mask" from SU because not everybody uses KT. (i will add it though)
-
That's true. And thanks for the tut!
-
at last some renderings...
pure kerkythea:
with an artistic photoshop filter:
another pure kerky:
KT + PS:
This last one is my first attempt to put a model in a photo (see original photo)
-
Hello Lucius,
Great model with lots of detail!
May I ask where you got the measurements of the aqueduct from? I have to make a 3D model of the pont du gard for a school project and while I can find some measurements online (like the height of the different tiers) I can't find all of them (like the height of the arch itself, or the position of the middle tier on the bottom tier). Also, the information I found is often inaccurate. It would be really helpful if you could tell me where you got the measurements from.
Thank you in advance!
Advertisement