Perspective studies from photo's: Vanishing points
Checking a thread from the environmenting subforum, I made some perspective studies from photo's my self. They're mostly from street allies since they have mostly more vanishing points because they're not alligned straight to each other.
I wonder, in some of these, the VP's are seperated from each other. Is this because of me or are the buildings really skewed? I wonder if the cube in the first street is correct. Some advice and if nescesary paintover would be apreciated.
Looks like it's just that each of the building is not parallel to each other so they have different vanishing point. The cube you did on the first one looks skewed because it uses the vanishing point of two buildings whereas the last one you use just the left building's vanishing points.
The third one looks odd too because I don't think you can have 2 horizon lines at once. I guess the photograph used special lens or something since the bicycles and the figure recedes back very dramatically on a fairly short distance but I'm just guessing.
You can't have two horizon lines because you can't have two horizons as long as you're on a planet. But you can have vanishing points that sit off the horizon line, such as a street going up or down a hill. Further, if the slope of the hill is not constant, the vanishing point can move further or closer to the horizon line for parts of the street further along the hill.
I would also agree that some of the photos might have some lens distortion, which could be causing some of the trouble. You really have no idea for sure if there's camera distortion or how much- which is one of the reasons it's best to shoot your own reference when possible so that you actually see the subject in person and can supplement the photo with your firsthand experience.
But enrigo is right that a lot if it is likely because the buildings are not necessarily parallel to each other.