Re: [css3-color] ICC implementation

On Aug 31, 2008, at 7:22 AM, David Woolley wrote:
>
> The assertion was that camera manufacturers were routinely not tone  
> mapping at all, not that they were doing it wrongly.  You more or  
> less accept this yourself, by saying that professional cameras  
> capture the raw readings.

All digital cameras capture "Raw" in the sense that this is  
essentially raw sensor data. All have to do this. Consumer cameras  
that don't give the end user access to this as a Raw file (or DNG),  
perform their own in-camera JPEG conversion and supply the user with  
only a JPEG. It is a function of the in-camera JPEG conversion to tone  
map the image. All cameras producing a JPEG do this.

So again the assertion that camera manufacturers routinely do not tone  
map the date when producing a JPEG is absurd on its face because a  
linear capture JPEG is a.) a departure from the JPEG spec, b.) a  
departure from the EXIF spec, c.) probably a departure of the ICC  
spec, the sRGB spec, and possibly other specifications as well. But  
most importantly, the image would look like complete crap. Anyone who  
is not blind would return the camera. The camera would fail in the  
market place. What you are suggestion is a camera that would produce  
such bad results it would not be correctable and remain acceptable.

> The problem is that a lot of users don't realise that they then need  
> to apply this correction before putting the image on a web site.   
> They assume the camera produces an image that already complies with  
> the web standards.


It's completely beyond absurd to state categorically that tone mapping  
isn't being done with in-camera JPEG, and that in particular end-users  
would be expected to perform tone mapping themselves (how the f would  
they do this anyway in your word view?) This is just incredibly bizarre.

Every single frickin image that's taken with a digital camera and  
produces a JPEG explicitly states that it does comply with web  
standards, that it has been tone mapped into sRGB.

> It may well turn out that the problem images come from would be  
> professionals who are using cameras that are too sophisticated for  
> their level of knowledge, but such images are quite common.

And yet you have provided exactly zero examples of images that clearly  
exhibit this phenomenon with demonstrable data showing that the scene  
was correctly exposed. What you have provided thus far are images that  
almost certainly underexposed (hard to say with metadata stripped).

>> adjusting either white point, or midtone in the examples you  
>> supplied, the shadows are very noisy which is a classic case of  
>> underexposure, not improper tone mapping.
>> Further the suggestion that the "example of images that are clearly  
>> gamma 1.0" makes zero sense. Tone mapping that would result in an  
>> image TRC defined by gamma 1.0 would not make it out of R&D let  
>> alone for
>
> But youi said that that is exactly what professional cameras do.   
> Given that sensor count photons, the natural response of a camera is  
> gamma 1.0.

The in-camera JPEG is tone mapped. It's only the Raw that is linear  
encoded. For a consumer camera, the end user doesn't have access to  
that file. For a professional, the Raw file must be processed with a  
Raw converter, which has a primary function of tone mapping images.

Chris Murphy

Received on Monday, 15 September 2008 16:53:05 UTC