Ok this function is really super cool to think about. One fun thing I thought about is that it not only represents everything we can see visually but could even be thought of as some sort of memory indexing function. In a way, we use this in our brain to call up specific points in time and space on demand that are tagged with smell and sound files.
frankieeder
Does this mean that any attempt at recording our reality is a sampling of this plenoptic function, that also contributes to the plenoptic function once distributed?
fywu85
Does this account for the fact that some people have color blindness / weakness? Also, this seems not to count the biological nature of the eyes: a bright room we see when we just wake up from sleep is different from the one when our eyes have adapted the brightness.
youtuyy
In short, the function captures 3D position, 2D direction, color and time. It basically has everything we need. I think that if we can also use it as a light field function with certain assumptions. First, we do not need to worry about time and color. And we can also assume that light is constant, which can be represented as a 2D direction instead of 3D position.
afang-story
@fywu85 I believe this does take into account for color blindness because the common example of color blindness is only having two types of cones, which just means that they will perceive the given wavelength differently than those who have three cones, and wavelength is one of the variables of the function
drewkaul
It's interesting to note that environment mapping, which we did in some of the earlier projects, is related to the plenoptic function. Reducing the plenoptic function from a 5d function to a 2d function by constructing it for a particular point is analogous to representing a view of an environment from a fixed point as a 2d texture map.
Ok this function is really super cool to think about. One fun thing I thought about is that it not only represents everything we can see visually but could even be thought of as some sort of memory indexing function. In a way, we use this in our brain to call up specific points in time and space on demand that are tagged with smell and sound files.
Does this mean that any attempt at recording our reality is a sampling of this plenoptic function, that also contributes to the plenoptic function once distributed?
Does this account for the fact that some people have color blindness / weakness? Also, this seems not to count the biological nature of the eyes: a bright room we see when we just wake up from sleep is different from the one when our eyes have adapted the brightness.
In short, the function captures 3D position, 2D direction, color and time. It basically has everything we need. I think that if we can also use it as a light field function with certain assumptions. First, we do not need to worry about time and color. And we can also assume that light is constant, which can be represented as a 2D direction instead of 3D position.
@fywu85 I believe this does take into account for color blindness because the common example of color blindness is only having two types of cones, which just means that they will perceive the given wavelength differently than those who have three cones, and wavelength is one of the variables of the function
It's interesting to note that environment mapping, which we did in some of the earlier projects, is related to the plenoptic function. Reducing the plenoptic function from a 5d function to a 2d function by constructing it for a particular point is analogous to representing a view of an environment from a fixed point as a 2d texture map.