Keeping track of the new vocab in this course is essential to my understanding, so let me share what the terms here mean:
View Plane: Theoretical plane where the projected image is formed. Think of it as the virtual screen where the rendered scene gets flattened.
Near Plane: The near plane is like the glass pane closest to you. Objects closer than the near plane are not rendered.
Far Plane: Maximum visible distance. Objects beyond the far plane are also not rendered, as if they are beyond the horizon.
caelinsutch
I'm curious if the theory behind deciding this was how perspective would be represented computationally was derived from photography - or was there a more foundational theory behind viewing perspectives? Like why represent computer graphics as volume based "planes" when most other UIs are represented as "stacks"
ttalati
I am a bit confused about when each of these transforms is applied in the overall pipeline. Was the projective transform introduced just to motivate what goes on in the overall perspective transform of where we try to project onto a 2x2x2 cube. Or is a perspective transform first applied and then to project onto the actual screen a projective transform is applied--I am thinking the latter is not true since a screen transform seems different from a projective transform as the screen may distort the axis due to different aspect ratios?
grafour
Perspective projection: a mapping from three dimensions onto two dimensions
Keeping track of the new vocab in this course is essential to my understanding, so let me share what the terms here mean:
View Plane: Theoretical plane where the projected image is formed. Think of it as the virtual screen where the rendered scene gets flattened.
Near Plane: The near plane is like the glass pane closest to you. Objects closer than the near plane are not rendered.
Far Plane: Maximum visible distance. Objects beyond the far plane are also not rendered, as if they are beyond the horizon.
I'm curious if the theory behind deciding this was how perspective would be represented computationally was derived from photography - or was there a more foundational theory behind viewing perspectives? Like why represent computer graphics as volume based "planes" when most other UIs are represented as "stacks"
I am a bit confused about when each of these transforms is applied in the overall pipeline. Was the projective transform introduced just to motivate what goes on in the overall perspective transform of where we try to project onto a 2x2x2 cube. Or is a perspective transform first applied and then to project onto the actual screen a projective transform is applied--I am thinking the latter is not true since a screen transform seems different from a projective transform as the screen may distort the axis due to different aspect ratios?
Perspective projection: a mapping from three dimensions onto two dimensions