I’d start with 3 vans sellect all 3 layers command T (transform) and use the perspective. or you could use vanishing point.
to get that right, you are going to have to shoot perspective lines off the key elements of the front van (or another of your choosing) making a perspective grid, and aligning the vans in the receding positions to the receding perspective lines.
The perspective in your original photo will by definition only be correct for one van. I don’t know how successful this is going to be.
Jean:
Vanishing Point’s Clone tool will do exactly that for him automatically.
Tip: do the cloning in VP on empty overlying Layers.
I’d like to see that done. I’ve never seen Vanishing Point used for this sort of a project.
One of the on-line demos of VP did something similar: (adding windows in perspective along the receding wall of a Venetian palace) if I recall correctly.
But try it for yourself and see.
I have used VP to create extra flooring and patterned carpet; and decorative plaster ceilings; in an interior shot which had had to be taken with a fixed-back camera and then corrected for converging verticals in VP thus losing the ability to use the full width of the image.
I remember posting the Before-and-After results of that one here (plus, I think, Screen shots of the VP window) but it was some years ago now and will long-since have disappeared into the archives.
what you’re describing is much simpler task than what fotografica needs to accomplish.
No, it is exactly the same thing. How much have you experimented with VP because, with a little practice, it can be a very powerful tool.
He needs to clone his single van and then "paint" the Clone on an in-perspective grid on successive layers.
Then mask each of those layers individually, as necessary, to hide any parts of each van that may need to be hidden.
As I said before: Try it for yourself and see what works for you but VP is the tool that I would use to tackle this task.
Each face of the van would have to have it’s own vanishing point plane drawn for it. The top of the van, the front, the hood, the windshield, etc. That’s a lot of work, and then having to clone all those parts and have the different-planed faces match up would be more work than I’d be willing to do. Besides that, for realistic perspective, you would have to fake in things like how the furthest van would have that one headlight hidden, but the one closest to you should have both headlights visible. You can’t really make 3d out of 2d like that, and I have a feeling that faking something only partway decent would be a lot of work.
I imagine that the six vans could be "parallel parked" like my Henry Moores?
If not, wouldn’t it be a lot quicker, easier (and therefore cheaper) to just retake the photograph with all six vans in the shot?
And what is this for? If it’s for the web, it’s only going to receive the quickest of glances anyway.
Wow!
Thanks for all the advice I will try Ann’s suggestions using VP. Shooting the 6 vans is not an option, so I will work on it and post my work as I progress. By the way fotografica is the name of the studio! My name is Ana
The important thing when using VP is the "plane" or view-point on which you base your grid.
For the Henry Moores, I used a this diagonally receding plane. The mesh needs to cover both the object that will be cloned as well as the extended area in which you will place the copy.
You will have to do each one separately on its own layer and then stack them from the hind-most forwards. Then make a silhouetted copy of the front van and put it at the top of
the layer stack.
This was the VP mesh that I used for the Moores:
<
http://www.pixentral.com/show.php?picture=1H1ydDTaBjWmBrIbDe BXaBibl4vq51>
Anna, I hope it works for you, but I remain skeptical. I agree with PeterK’s assessment.
Ann has some significant perspective errors in her test, which would not show so much on an amorphous Henry Moore sculpture, but are likely to be quite obvious on a highly structured subject such as a vehicle.
Your van has a lot more convergence going on than the sculpture does, has less structure to key off of (e.g., the base of the sculpture) and you need to cover a lot more space as well.
If you cannot reshoot all 6 vans together, is it possible to reshoot some of them, parking in different spots as required for the assemblage, and stripping the configuration together?
I suspect that you are making this a lot more complicated than either the project warrants or the budget can support?
8/
post a pic of the one van
yes, I have no idea what the assignment is beyond what has been stated. I too assume there isn’t much of a budget, or they would just shoot what they need.
My comments are on what is "correct" perspective, and how distorted and inaccurate I
think
the results will be with vanishing point. I don’t know, and will be interested to see what works out for Ana.
Personally, if the budget were low and the requirements modest, I’d render it in 3D, but I’m an illustrator, not a photographer, and I despise doing extensive retouching in Photoshop. So that is indeed my bias.
Buko,
How can I attach an image to this message?
the same way you posted the first image
Ana, the best way might be to post a PSD layered file (that you have zipped) on your web site, and give us the link. One van on a layer with transparency would be ideal.
