Skip to content

Commit

Permalink
add carrots image and link
Browse files Browse the repository at this point in the history
Signed-off-by: Peter Hillman <[email protected]>
  • Loading branch information
peterhillman committed Mar 23, 2024
1 parent dd63614 commit 6e737df
Show file tree
Hide file tree
Showing 5 changed files with 134 additions and 56 deletions.
17 changes: 9 additions & 8 deletions website/SceneLinear.rst
Original file line number Diff line number Diff line change
Expand Up @@ -11,12 +11,12 @@ proportional to the amount of light they represent. Most image formats are
display-referred, where the values indicate how much light should be used to
display the image on a screen (or how much ink to use to print the image onto
paper), and many image formats apply an encoding to the image so that the
numbers are not linear. Some sources use the term 'input-referred' and
'output-referred' as alternatives to scene-referred and display-referred
numbers are not linear. Some sources use the term "input-referred" and
"output-referred" as alternatives to "scene-referred" and "display-referred"
[#fterms]_.

This is a brief description of the difference between scene-referred and
display-refrred representations, what linear-light means, and why using
display-referred representations, what linear-light means, and why using
scene-referred images may be beneficial. For simplicity, the explanation here
assumes the image is photographed by a digital camera and is displayed on a
computer monitor. This discussion also applies to software generated images
Expand Down Expand Up @@ -81,8 +81,9 @@ the colorfulness of bright areas, and increasing the colorfulness of dark ones.
Such film-based images are so common that an image which has not had that
process applied appears incorrect, so many digital tone mapping approaches are inspired
by reproducing what film would do, to make the image appear more familiar and so
more natural. The images above illustrate this. The first version of the image
presents the pixels with the same color and intensity that the real objects had
more natural. The images above illustrate this. The first image is the
:doc:`test_images/ScanLines/Carrots` test image
processed so the pixels have the same color and intensity that the real objects had
[#fsrgb]_. This particular scene has low enough contrast and subtle enough colors
that monitors are capable of displaying it faithfully. The second image is a
JPEG created by the camera using its built-in tone mapping of the image. This
Expand Down Expand Up @@ -125,7 +126,7 @@ combined image appear as if a camera really photographed it, so we need to
compute what light would have been captured by the camera. This task is much
easier if the photographed background is a scene-referred image, and the
software that generates the character also stores a scene-referred image as seen
by the 'virtual camera' in the software. If both the photograph of the
by the "virtual camera" in the software. If both the photograph of the
background and the computer generated image are display-referred, with some
artistic look applied, it will be much harder to produce a convincing composite
image. OpenEXR images are widely used for motion picture visual effects
Expand Down Expand Up @@ -253,15 +254,15 @@ OpenColorIO.
.. [#ftonemap] Other terms are used for tone mapping and gamut compression,
such as applying a Rendering Transform, Look, Display LUT, Filter, or OOTF
(Opto-Optical Transfer function); many approaches combine both color and
contrast modification into a single step. Here, the term 'tone mapping' is used.
contrast modification into a single step. Here, the term "tone mapping" is used.
.. [#fsrgb] The colors in the first image will be rendered as observed on
monitors which are calibrated for sRGB with a maximum intensity of
62 cd/m :superscript:`2`. Most monitors will display the colors accurately
but slightly too bright.
.. [#feotf] Other terms for EOTF include applying a gamma or log function,
or simply a 'monitor curve'.
or simply a "monitor curve".
.. [#fscenelinear] One convention is to use the term *input-referred linear* for any
image where the values are proportional to how much light the camera captured,
Expand Down
1 change: 1 addition & 0 deletions website/test_images.txt
Original file line number Diff line number Diff line change
Expand Up @@ -14,6 +14,7 @@ ScanLines/README.rst
ScanLines/Blobbies.exr
ScanLines/CandleGlass.exr
ScanLines/Cannon.exr
ScanLines/Carrots.exr
ScanLines/Desk.exr
ScanLines/MtTamWest.exr
ScanLines/PrismsLenses.exr
Expand Down
56 changes: 56 additions & 0 deletions website/test_images/ScanLines/Carrots.rst
Original file line number Diff line number Diff line change
@@ -0,0 +1,56 @@
..
SPDX-License-Identifier: BSD-3-Clause
Copyright Contributors to the OpenEXR Project.
Carrots.exr
###########

:download:`https://raw.githubusercontent.com/AcademySoftwareFoundation/openexr-images/main/ScanLines/Carrots.exr<https://raw.githubusercontent.com/AcademySoftwareFoundation/openexr-images/main/ScanLines/Carrots.exr>`

.. image:: https://raw.githubusercontent.com/AcademySoftwareFoundation/openexr-images/main/ScanLines/Carrots.jpg
:target: https://raw.githubusercontent.com/AcademySoftwareFoundation/openexr-images/main/ScanLines/Carrots.exr


Stored with ACES2065-1 (AP0) chromaticities

.. list-table::
:align: left

* - lensModel
- 70-300mm F4.5-5.6 G SSM
* - channels
- A, B, G, R
* - expTime
- 0.625
* - lineOrder
- increasing y
* - displayWindow
- (0 0) - (599 399)
* - effectiveFocalLength
- 120
* - cameraModel
- SLT-A99
* - dataWindow
- (0 0) - (599 399)
* - chromaticities
- red (0.7347 0.2653), green (0 1), blue (0.0001 -0.077), white (0.32168 0.33767)
* - pixelAspectRatio
- 1
* - type
- scanlineimage
* - screenWindowWidth
- 1
* - screenWindowCenter
- (0 0)
* - adoptedNeutral
- (0.32168 0.33767)
* - compression
- zip
* - cameraLabel
- Sony SLT-A99
* - aperture
- 8
* - cameraMake
- Sony
* - isoSpeed
- 100
Loading

0 comments on commit 6e737df

Please sign in to comment.