"Tone mapping for high dynamic range images"
Originally Posted by ChrisWiggles
I can't speak for HoustonHoya, but I doubt that is a typo. That is, in essence, the basic claim that started this thread.
The essence of some of the claims in the thread is that we cannot see more than a few hundred to one across a single scene. Therefore displays with simultaneous CR performance greater than say 200:1 or so are unnecessary because they already exceed our visual capabilities. There have been a variety of specific numbers suggested that follow that basic claim, 100:1, 130:1, 200:1, 250:1, 300:1. Maybe I'm missing some. But they're all a few hundred to one basically.
It turns out that the ISF is not actually stating this at all and is instead misinterpretation by some dealers who were ISF-trained, and Silver instead cited Poynton's figure of 1,000:1 simultaneous capability across a scene. This is an order of magnitude lower than the 10^5 figure from the Brightside papers which seem more accurate to me intuitively. However, despite the fact that the original post of the thread which seemed to insinuate that the ISF was unfortunately parroting this low 200:1 capability figure was actually not accurate in terms of what the ISF is saying, clearly many people have come to try to defend these low figures as correct. At least for myself, it's very difficult to take such low claims as 200:1 across a scene as the max human capability seriously, but apparently this is fairly commonly held belief. Which is why this thread is already 9-pages long. I am not sure how people are capable of driving around at night with only 200:1 capability, yet somehow...
is a doctoral dissertation that says (pp. 3-4):
The dynamic ranges of several natural scenes were measured and 1:160 was found to be an average contrast ratio (Hunt 1995). Outdoors scenes usually have a larger dynamic range, which can reach a contrast ratio of three orders of magnitude (1:1000) or more. Scenes with fog tend to have a small contrast ratio.
1:1000 or more for outdoors scenes ... that seems to confirm Poynton and refute the idea that the human visual system (HVS) is limited to 1:300 or less. But, as what I am about to say shows, I think the dynamic range capability of the HVS is actually irrelevant.
Folks, look at it this way. If the HVS is
limited to a CR of 1:300 or less, then it is intrinsically "low dynamic range" (LDR). If that is so, then the dissertation I just quoted is pointless. There would be no point in developing high dynamic range (HDR) displays for LDR eyeballs.
I believe a source of confusion on this topic is the fact that a "standard display" is LDR. That is, until very recently all TVs and projectors CRTs, LCDs, etc. were LDR. This indeed is an assumption Poynton makes. I am sure it is the assumption the Imaging Science Foundation makes. It is the assumption made by the ATSC and NTSC television broadcasting standards, by Rec. 709 and Rec. 601, etc.
If the display is, by common assumption, LDR, then the signals that are input to it must be encoded accordingly. It's no good inputting a 1:1000 image into a 1:100 display. That would provoke clipping of whites and/or swallowing of shadow detail.
So the dynamic range of the image is reduced in the video camera from what it typically is in the original scene. If the original scene is an HDR one, taken in broad daylight, DR compression is needed. I assume that the same
DR compression has to be done even for LDR scenes as well, for the sake of consistency.
I further assume that even if the video camera or other capture device is inherently HDR, the DR of the output image still has to be reduced for an LDR display.
And I further
assume that the above is true even if the display is HDR
. That is, even though we now have displays and projectors capable of simultaneous contrast ratios well in excess of 1000:1, the video system as a whole can't assume that. It has to assume the end user's display is LDR, since that indeed is the assumption under which the relevant standards were developed.
So it seems fair to summarize the above as: the video systems we have today (ATSC, NTSC, Rec. 709, Rec. 601, etc.) are inherently LDR. Ergo, signals input to the TV are without fail optimized for LDR displays ... even if the display in "my" home theater just happens to be HDR. Consequently, the display will perform its best with these input signals when it is adjusted to reproduce an image using a low dynamic range.
Notice that the question of the actual dynamic range capability of the human visual system completely drops out of the above discussion. It is irrelevant.