Simulating human visual system, non-pov (25k) (Message 1 to 6 of 6)
From: Kari Kivisalo
Subject: Simulating human visual system, non-pov (25k)
Date: 20 Sep 2000 20:47:35
Message: <39C95AEC.20DB166A@kivisalo.net>
A couple of researchers at NASA published this post processing method
in IEEE TRANSACTIONS ON IMAGE PROCESSING.
A Multiscale Retinex for Bridging the Gap Between
Color Images and the Human Observation of Scenes
It's purpose is to visually match an image on low contrast rate display
to the high contrast rate real scene recorded at high bit resolution.
One interesting property of this method is that the physical image size
affects the processing.
This is way better than emulating photographic compression and it should
be automatic too.
______________________________________________________________________
Kari Kivisalo http://www.kivisalo.net
From: ryan constantine
Subject: Re: Simulating human visual system, non-pov (25k)
Date: 20 Sep 2000 21:52:16
Message: <39C969E2.706EA59A@yahoo.com>
or the ceiling could be the ever hated cottage cheese texture they used
to use. thank goodness ceilings are now textured the same as walls
these days.
"Tony[B]" wrote:
> > Hmm... the roof looks splotchy. Is that a problem with the method or just> the JPEG?
From: ryan constantine
Subject: Re: Simulating human visual system, non-pov (25k)
Date: 20 Sep 2000 21:52:34
Message: <39C969F5.BDDF47C0@yahoo.com>
most fascinating.
Kari Kivisalo wrote:
> > A couple of researchers at NASA published this post processing method> in IEEE TRANSACTIONS ON IMAGE PROCESSING.> > A Multiscale Retinex for Bridging the Gap Between> Color Images and the Human Observation of Scenes> > It's purpose is to visually match an image on low contrast rate display> to the high contrast rate real scene recorded at high bit resolution.> One interesting property of this method is that the physical image size> affects the processing.> > This is way better than emulating photographic compression and it should> be automatic too.> > ______________________________________________________________________> Kari Kivisalo http://www.kivisalo.net> > ------------------------------------------------------------------------> [Image]
From: Kari Kivisalo
Subject: Re: Simulating human visual system, non-pov (25k)
Date: 20 Sep 2000 21:56:45
Message: <39C96B24.3CFC55C0@kivisalo.net>
"Tony[B]" wrote:
> > Hmm... the roof looks splotchy.
This was 1996-97. There were commercial 12bit cameras at that time
but since NASA put all the money on the Mars probes these rocket
scientists had to use 8bit cameras. That is black level noise which
is amplified in the process.
______________________________________________________________________
Kari Kivisalo http://www.kivisalo.net
From: Bob Hughes
Subject: Re: Simulating human visual system, non-pov (25k)
Date: 20 Sep 2000 22:45:09
Message: <39c97635@news.povray.org>
I'd expect to see the room interior bright enough to see okay but then not a
brighter lamp and window (outside scenery). This shows it somewhat overall
lightened.
Bob
