VI and Camera Placement (for Devin "Maxer")

[Virgil Johnson]

Devin

 

This comes from the perspective of a visual impact study and may be way overkill on the accuracy level.

But as I have said in visual impact nothing is left to chance.

 

Firstly one must always be sure of the placement of cameras.

You have to know exactly were you are in absolute and relative terms.

 

You use maps. DOT maps. USGS maps. Topo Maps.

GPS. You measure in the field from known and located objects. (Be sure to have the surveyor locate everything that you may need.)

 

Know the AMSL of your camera and objects if relevant.

Know the relative difference between the camera heights and your objects.

 

Know the azimuth and tilt of your camera.

 

Know the difference between magnetic and truth north when needed and where magnetic north was the day you photographed.

 

If you do not locate the camera correctly in your model the scale and position and rotation will be off and your study will be bogus.

Anyone reviewing this study for the opposition will have you.

(And yes people opposing projects are getting savvy about visual impact studies.)

 

You must also locate the objects in your photo and mark the accordingly. Poles, flags, balloons,

etc. at selected (and surveyed?) points.

Be sure you know what is what because you will be drawing references in your model and using them for placement.

 

Get ready for lens distortion and prepare for it in selecting where you put your markers.

 

For visual impact studies be sure you are putting the camera in the right location and you know what is in the FOV. You are limited (except in rare cases) to 50mm shots - standard for the human eye - and have to work within this limitation.

And the photo that you get will be critical to your study so make sure it is right.

 

Photos are everything in the visual impact studies. (The last animation I did was a walk through of a sound studio in Kula Lumpur in the summer of 1994.)

So lens theory is important to learn.

 

Hope this is somewhat of a help.

 

Virgil

[Noise]

Very interesting stuff Virgil.

 

I have recently got interested in making the camera match process more accurate and less of a tweakathon. Although, I am not sure I want to get into the area you are in as it is heavily techincal and I am more art based.

 

In the UK we call your process, Verifiable images and I think it was Hayes Davidson that had a strong part in developing this as a standard process. They have a book called Accurate on it below;

 

http://www.hayesdavidson.com/html/publications/publications.html

 

Its good to have experienced sources for this type of work as I think it will become much more prevalent in the near future especially as the law gets out of control.