the vanishing points meet at the station point at a right angle, correct? but there are many ways for that to be true, as seen in pic related. so how do you place the station point correctly in an established scene?
slide thread
>>7860507how do you mean?
>>7860511>station point is useless after vanishing points are set>determining it solely from vanishing points is impossible>bickering about perspective and drawing lines with a ruler is an enjoyable pastime for nodraw dunningkrugersso you're either fishing for nodraw dunningkrugers are one of them
>>7860504>the vanishing points meet at the station point at a right angle, correct?Correct, but the station point is always on the center line, so there is only one possible solution for any two vanishing points.
>>7860525
>>7860540so let's assume for a second that the third vanishing point is far enough to ignore and we knew where the centerline was even though I mentioned it's impossible with vanishing points only... what is that cool useful thing you can do with this now?
>>7860543Well the center line is always in the center, hence the name. It's why I made the little X on the photo.And you can use your station point to find the location of vanishing points for angles other than your two 90 degree ones you started with. So if you wanted to put a car in the middle of the intersection rotated about 30 degrees from the street, you can measure a 30 degree angle from the street's vanishing point and find where it should go.
>>7860551>Well the center line is always in the centernope, there are these things called shift lenses.>And you can use your station point to find the location of vanishing points for angles other than your two 90 degree ones you started withooh, so exciting, geometrically accurate _rotated_ boxes in two point perspective, that will sure come in handy!
>>7860527but how does one determine the center line in the pic in op?>>7860540i dont think the center line is always in the center of the picture plane though. you can crop photos and drawings, meaning the center point isnt always reliably fixed.>>7860543station point lets you measure things accurately. if thats not something you care about, then no, you dont need it.
>>7860559>measure things accuratelycan you show what the fuck you mean?
>>7860559It’s always the line that goes through the center of the picture plane and is perpendicular to the horizon. The picture plane is not the canvas. If you cropped, then good luck, just guess.
>>7860562>good luck, just guessLOLso much for "accuracy"
>>7860562>The picture plane is not the canvas. oh, good point. just seems odd to me that such an important thing as the sp cant be worked back to.
>>7860504Why is a station point important when it's always in the middle
>>7860577it is wewy impowtant if you wanna dwaw a cube rwotwated at anowew angew in a two pwoint pewspewctiwe
>>7860579Ok? Why not just draw the horizon line slanted
>>7860581nononoFor example you have a box facing N and you want a box facing NNE. Then you fuck around with the station point, since you can draw arbitrary additional right angle triangles, et viola, a girl with pe- I mean another pair of VPs.
>>7860577the station point is where youre standing in relation to the picture plane. it determines stuff like center of vision, field of view, and lets you measure lines that recede in space.
>>7860590>is where youre standingWrong, it's a point in an overlaid top-down projection of the scene. You can actually move that shit further away if you don't want to crowd your canvas with retarded lines.
>>7860504This thread is the blind leading the blind. Both of those SPs can be correct. The SP will correspond, horizontally, with the central point of vision on the horizon line; this is where all the lines in a one-point perspective will converge. Artists DO NOT always place this at the center of the picture.
>>7860540Hello NotFuji, I hope you're having a good day today buddyCute character . :-]
>>7860658so how does one find it in a scene such as this? >>7860540its not where that anon placed it, we can tell from the pavement lines, right?
>>7860540That's rightMind you, this only works if you're reverse engineering an object with 45° corners
>>7860853>anon is non-euclidean
>>7860821It's a bad photo for studying perspective, because the lines of the left side of the building converge at a different point on the horizon than the lines on the right side of the building. It could be that the street is on an incline, i.e. going uphill towards the right.
>>7860552>that will sure come in handyokay, it won't, you're right, stop drawing and post more
>>7860871Whoops, I meant 90° corners.
>>7860821>>7860881I don't see what you mean. The camera is just tilted. All the lines line up.>>7860853Technically you can do it with any angle, you just have to know the angle. Instead of the station point being a right angle it will be whatever angle your known VPs are at.>>7860574I'm sure you can, I just don't know how and don't feel like finding out right now-Nevermind I just figured it out, if you have a third vanishing point like a diagonal on a square, there is only one possible station point from those three points, and your center line will be on that point as usual.
>>7860917>I don't see what you meanThe horizon line is pretty obviously lower on the left side of the building than it is on the right side. On the far left it aligns with where the wall meets the foundation, and on the far right side it's like 30cm above the foundation.If you assume that the camera is at eye level, and that the doors of the building are at least two meters tall, which is legally mandated all over the world, then the point of the foundation line closest to the viewer should be lowest in the picture relative to the horizon, and both of the far corners should be higher. It's pretty obvious that the road going to the right is going uphill. Obviously the road has a vanishing point higher than the water level horizon.
>>7860898oké>>7860917>if you have a third vanishing pointif you have three point you're fucked regardless because vanishing points travel in a hyperbola when box is rotated around one of its axes.
>>7860917Retard
>>7860926My lines are probably just a couple MOA off. Or it is on a hill. You're pretty unlikely to get an accurate answer from a blurry photo like this. But it's still close enough.>>7860932Vanishing points don't all have to be 90 degrees from eachother, anon.>>7860934no u
>>7860939I'd like to ask you to briefly think what happens when the square with dashes on its edges rotates and where 1 and 3 end up
The diagram you're building here isn't the same view as the actual scene. For a perpendicular vanishing point pair on a horizontal line, the station point diagram is actually a top-down isometric view of the station point and picture plane. You're just overlaying it on the projection because the measurements on that horizon line match between the diagram and the projection. If, like in >>7860540, your perpendicular vanishing points aren't on a horizontal line, the station point in the diagram you draw needs to be on a line extending from the image center in a perpendicular direction to the line connecting the vanishing points, not just straight down.
>>7860945Just rotate the big square around the SP. EZPZ
>>7860917>Technically you can do it with any angle, you just have to know the angle.If you know the angle then there's no point to reverse engineering because you'd be an omniscient god.
>>7860964>anon doesn't know how to use a map
>>7860917>I don't see what you mean. The camera is just tilted. All the lines line up.the pavement stones provide a rough line toward the center of vision, which seems to be to the right of that placement. that should be true even if the road is inclined iirc.>if you have a third vanishing point like a diagonal on a squarebut you wouldnt know its a square unless you have the sp to create measuring points with.
>>7861197>the pavement stonesThat would only be the case if the camera were facing the same direction as the stones. The camera is facing slightly more toward the front of the building than the side.>but you wouldnt know its a square True, but ideally you'd have an object in the scene that you do know the rough dimensions of so you can find a diagonal. It doesn't necessarily need to be square. But you do need that third point either way. Even in the ideal case if your center line is actually in the center it's acting as the 0 degree vanishing point.
>>7861219see pic related for what i mean. regardless of how the stones are facing they make a pattern you can use to trace their convergence in that shot to a rough area, and it's not where the other anon placed the center of vision.
>>7861247Why would the stones converge toward the center of vision? It converges toward it's vanishing point, like everything else.
>>7861249hm, yeah, dumb assumption on my part.
How to use any of these nonsense in mine then. The fun is, very hard so... Is it, the more-god, then?
