Using References Drawing Drawn or Photographed At An Angle

Albert wrote in news:1194662504.649578.200610 @v2g2000hsf.googlegroups.com:
I recall a computer program that was developed for making 3-view drawings using scans of articles - I had a demo of it somewhere, once. If I can find it, I'll post the name here and let you know. But don't hold your breath.
RobG (The Aussie one)

That IS indeed a tough one. When I am taking my own pictures, I can ensure that I am perpendicular to major axes.
However, on some subjects I am left with only quartering pictures. If one were to use pure photogrammetry, one needs to know the distance from the subject to the camera as well as the camera lens focal length, not very practical unless you know the photographer.
Therefore, you can only deal with relative distances. You need some basic dimensions that you will NOT be able to obtain from the photos.
Vertical dimensions are the easiest ones to scale, since most prototypes are shorter in that direction than in other directions. You assume that everything in a vertical column is indeed vertical and so is the same distance from camera (assuming camera is held reasonably level). Now you can measure distances along that vertical and ratio the sizes. This does only give you relative dimensions, which is why you need some basic dimensions.
While many photo editing programs these days have a perspective correction function, this is certainly imperfect for any direction with a component along the camera line of sight.
BTW, I taught myself to use CAD when I was trying to make scale drawings from a number of photographs of a race car. Fortunately, I did know the wheelbase and width of the car, and the tire sizes. I used what I learned from that exercise to take or seek out photographs of many vintage race cars, which I did five-views of, and I do sell a CD-ROM of scale drawings and photos for vintage oval track race cars.