Corbis Technical Guidelines: Commercial and Editorial Photography
1. Image Settings
1.1
1.2
1.3
1.4
Naming Conventions
Setting Up Your File Saving Preference
Required File Size
Setting Up Color Settings
2. Processing files for digital submissions
2.1
2.2
2.3
2.4
2.5
2.6
Highlights and Shadows
Cropping
Posterization
Over Sharpening
No Extraneous Image Data
Interpolation
3. Metadata
3.1 IPTC Guidelines for Commercial Content
3.2 IPTC Guidelines for Commercial Editorial, Historical and Fine Art Content
3.3 How To Input Metadata Using Adobe Photoshop
4. Glossary
Corbis Technical Guidelines: Commercial and Editorial Photography - Page 1 of 22
1. Image Settings
1.1 Naming Conventions
Image names should be limited to 20 characters including the file extensions. File extensions should be
included (ex. .TIFF; .jpg). Note: Do not use spaces, dashes or any unusual characters such as “,%&!?/().
Such characters cause problems with internal Corbis’ systems.
1.2 Setting up Your File Saving Preference
When working on a Mac, you can automatically
add a file extension to your files through
Photoshop.
Under the File Handling Tab in your
Photoshop preferences, set the
“Append
File Extension” to “Ask When
Saving.”
1.3 Required File Size
Scanned File Size Requirements
The recommended final scanned image file size is 50 MB, 8 bit RGB TIFF in Adobe RGB (1998)-Color SpacH
Maximum file size is 120 MB. Images should be fully SURFHVVHG
Digital Camera File Size Requirements
The recommended image capture file size is 50 MB, 8 bit RGB TIFF in Adobe RGB (1998)-Color Space.
Maximum file size is 120 MB. Images should be fully processed.
If the original capture is not 50 MB then the file will need to be non-destructively interpolated to 50 MB
(please see 2.6. Interpolation / page 8). The best results for interpolation are obtained from professional
cameras with a minimum capture size of 11 megapixels (approximately 33 megabytes).
File Format Requirements
TIFF in Adobe RGB (1998) Color Space
Image Compression should be set to: None or LZW Compression
Byte Order should be set to: Macintosh
Preview File Requirements (low res digital files for Editing of larger submissions)
Preview files should be 1280 pixels on the largest dimension
Preview files must be saved as a JPEG with a compression setting of 9 or higher
Corbis Technical Guidelines: Commercial and Editorial Photography - Page 2 of 22
1.4 Setting Up Color Settings
The color space (Adobe 1998) is set in your Color Settings preference
This setting is accessed at the following locations:
PC: Edit Menu\Color Settings
OSX: Photoshop Menu\Color Settings
You set your Color Space by changing the
RGB working space to “Adobe RGB (1998)”
Corbis Technical Guidelines: Commercial and Editorial Photography - Page 3 of 22
2. Processing Files for Digital Submissions
2.1 Highlights and Shadows
Shadows and Highlights of an image should have some detail present, except in the case of a specular
highlight. RGB highlight and shadow values should be balanced and appropriate for the look and feel of
the image with maximum values of:
Black point can be down to 5,5,5, White point can be up to 250,250,250
If an image has excessive white or black areas then often these can be recovered when converting the
file. Image A has areas of white which are measuring 255 in all 3 channels, this can often be fixed by
moving the recovery slider, this should bring the values back to an acceptable level. Image B has areas of
extreme black which are measuring 0 in all 3 channels, this can often be recovered by reducing the values
on the black slider.
Corbis Technical Guidelines: Commercial and Editorial Photography - Page 4 of 22
Why compress the tonal range?
This is the best means to ensure that there is detail in both the highlights and shadows of any image.
What are spectral highlights and pure blacks?
Spectral highlights are areas that, in reality, are brilliant white. Pure blacks are just that, pure black.
Bringing a highlight down to 250 doesn’t do significant damage to an image file. Similarly, bringing a pure
black up to 5 doesn’t do any significant damage. The risk of compressing the total tonal range by 10
points, 5 at the low end and 5 at the high end, is preferred over getting whites and backs that should
retain detail blown out or blocked up.
Limiting blacks to 5,5,5 and whites to 250,250,250 also serves as a quality control tool for our clients.
When a client gets an image with these black and white points, they can be confident that all the
available data contained in a piece of film or transparency has been captured.
Corbis Technical Guidelines: Commercial and Editorial Photography - Page 5 of 22
2.2 Cropping
Crop out rebate edge on film and/or borders
This image is cropped incorrectly. The rebate
edge is still visible
This image is cropped correctly.
Note: When cropping an image, make sure that the “Width”, “Height” and “Resolution” are left blank, so
that you do not interpolate the file:
Corbis Technical Guidelines: Commercial and Editorial Photography - Page 6 of 22
2.3 Posterization
Poorly scanned, poor digital conversion of images or overly manipulated digital images can often display
posterization.
Un-manipulated file
Manipulated file. Some Color adjustments have
been made. Shadows are starting to posterize.
2.4 Over Sharpening
Do not sharpen your images in software. If overdone, sharpening introduces damaging artifacts.
Un-manipulated file
Manipulated file: Sharpening has been
added to the file. Eyelashes are starting
to “halo.”
Corbis Technical Guidelines: Commercial and Editorial Photography - Page 7 of 22
2.5 No extraneous image data
All images should be clear of any extraneous image data, such as Photoshop Guides, Paths, Vector Masks,
Extra Alpha Channels, Extra Layers and Adjustment Layers.
A Normal RGB file
File with extra
Alpha Channel
A Normal file with just
Background layer
No paths
With extra work
path and extra
vector mask
A File with
no Guides
File with extra layer
and extra
adjustment layer
A file with Guides, to
remove the Guides,
click “clear Guides”
Corbis Technical Guidelines: Commercial and Editorial Photography - Page 8 of 22
2.6 Interpolation
Using software to increase image files size can introduce damaging artifacts. Please refer to next page for
proper interpolation guidelines.
This image is a 1.29 MB scan displayed
at 100%
The same image interpolated up 700%
to 38.9 MB, also displayed at 100%
Recommended Interpolation Methods
1
There are two different interpolation methods that are recommended by Corbis : Adobe CS RAW
interpolation and Adobe Photoshop Bicubic Interpolation. In addition, there are other third-party raw
software and interpolation software that do a more than adequate job of interpolating files. If these
options are not available, check with your camera manufacturer for alternative software solutions to
enlarge your files. All interpolations must be done without damaging the image quality.
1
The screenshots and third party software applications referenced in this document are provided as an example only. Corbis
does not endorse or support any such third party software applications or protocols.
Corbis Technical Guidelines: Commercial and Editorial Photography - Page 9 of 22
Adobe Photoshop RAW Interpolation
The RAW format lets the photographer interpret the image data rather than letting the camera make the
adjustments and conversions. One of the simpler and better methods of interpolation is using Adobe
Photoshop CS’s Camera RAW.
1. Open the RAW file with Adobe Photoshop CS.
2. The native RAW file size is selected in the size menu.
3. Depending on which camera you are using, the pixel dimensions available will vary. Use the following
pixel dimensions as a guideline
5000x3400 – 50mb / 4120x2802 – 33mb / 2960x2015 – 17mb
4. Make other adjustments as needed to the various image settings, and then click “OK” at the top of the
window to process the RAW file.
This is our preferred method for interpolating images because interpolation occurs using the RAW data
from the camera, which produces better results than interpolating from an image that is in TIFF or
Photoshop format.
Corbis Technical Guidelines: Commercial and Editorial Photography - Page 10 of 22
Adobe Photoshop Bicubic Interpolation
The second option is to use Photoshop’s Image Size tool to interpolate the photo to the required size.
1. Open a file with Adobe Photoshop.
2. Choose Image Size... from the Image header.
3. Ensure the Constrain Proportions and Resample Image dialogue boxes are checked; and choose Bicubic
Smoother from the drop-down menu. The Bicubic Smoother algorithm is best used when making
enlargements; Bicubic Sharper is best used when making reductions.
4. Set your Resolution to 300ppi (pixels per inch), and enter a Width or Height value. You can confirm the
new file size by checking the Pixel Dimensions at the top of the dialogue box.
There are other methods of interpolation, such as stair stepping and interpolation software, however the
results must not degrade the file.
All interpolations must be checked to ensure there are no artifacts or aliasing. Interpolation must not be
so severe as to damage the file!
Corbis Technical Guidelines: Commercial and Editorial Photography - Page 11 of 22
3. Metadata
3.1 IPTC Guidelines for Commercial Content
The IPTC fields listed below are the only fields that Corbis automatically imports into our editing and
cataloging programs. You do not need to fill in any other fields for Corbis. Your images can have information
in other IPTC fields, but Corbis disregards that information. Corbis also retains the EXIF data for the digital
files you submit, which is the technical information written by the camera (ie. ISO, ƒ-stop, shutter speed).
1.
Headline: The Headline (referred to as the Title at Corbis) should provide the reader with a short and
precise explanation of what appears in each image. Photographers may add titles themselves. If not,
titles will be added during the keywording process. Note: Use the Headline field instead of the Document
Title field to add a title. The Document Title is not supported by the upload process.
2.
Description: The Description field should be used to enter any descriptive or helpful information about
the image. This information may include model ages, ethnicities and notes about any relevant
manipulation or compositing. Although it is not generally applicable for creative content, the Description
field may also be used to provide a Caption, in addition to the title, to add further explanation or
additional context. Note: It is not necessary to describe elements of the image that are obvious, or easily
understood by looking at the image.
3.
Keywords: Please provide keyword suggestions; especially for images that convey an abstract idea or
concept that may not be easily recognized in the image. Note: Similar to adding information to the
Description field, it is not necessary to add keywords that describe elements of the image that are
obvious, or easily understood.
4.
City / State / Country: Enter all location information into the appropriate fields.
5.
Date Created: Do not provide dates for creative content.
Corbis Technical Guidelines: Commercial and Editorial Photography - Page 12 of 22
3.2 IPTC Guideline for Editorial, Historical and Fine Art Content
The IPTC fields listed below are the only fields that Corbis automatically imports into our editing and
cataloging programs. You do not need to fill in any other fields for Corbis. Your images can have information
in other IPTC fields, but Corbis disregards that information. Corbis also retains the EXIF data for the digital
files you submit, which is the technical information written by the camera (ie. ISO, ƒ-stop, shutter speed).
1.
Author - Enter your name exactly as it should appear in the credit line on the Corbis website.
2.
Headline - The Headline (referred to as the Title at Corbis) should provide the reader with a short and
precise explanation of what appears in each image. Photographers may add titles themselves. Note: Use
the Headline field instead of the Document Title field to add a title. The Document Title is not supported
by the Corbis upload process.
3.
Description - The Description (referred to as the Caption at Corbis) should provide further explanation
and additional context – the who, what, where, when, why and how (where applicable) – for each image.
Use this field to record a time span or when you do not know the exact date Enter an time span or
approximate date photographed using one of the following formats: ca. May 1963; 1992-1994; early 1970s;
March-June 1956; February 2004; 1997; ca.1941; 1940s.
4.
Keywords - Please provide keywords. Keywords should describe what is literally depicted in the image
(man, dog, house) as well as abstract ideas or concepts that are conveyed by the visual elements (loyalty,
companionship, security). Do not over-keyword by describing minor or irrelevant details. Focus keywords
on the primary subject of the image.
5.
City / State / Province / Country - Enter all available location information into the appropriate fields.
Also enter this information as keywords.
6.
Date Created - Enter the exact date that the photograph was created.
Corbis Technical Guidelines: Commercial and Editorial Photography - Page 13 of 22
3.3 How To Input Metadata Using Adobe Photoshop
1. Open the file you wish to add metadata; and choose File Info... from the File header.
2. This will open the File Info dialog box.
3. Choose IPTC Content tab from the left panel. This is where you will enter metadata for Headline,
Description and Keywords.
Corbis Technical Guidelines: Commercial and Editorial Photography - Page 14 of 22
4. Choose IPTC Image tab from the left panel. This where you will enter metadata for Date Created,
City, State/Province and Country.
5. Once you’ve finished entering your metadata click OK, and choose Save from the File header.
Corbis Technical Guidelines: Commercial and Editorial Photography - Page 15 of 22
4. Glossary
Additive Color Theory
The principle of creating color from projected light. The presence of equal colors of all light creates
white (all light on); the absence of all light creates black (all light off). Monitors and scanners work in
additive color (RGB).
Additive Primaries
Red, Green, Blue. The three primary colors that, when combined in varying amounts, produce the
visible spectrum.
Alpha Channel
An image-editor channel used to contain a mask or partial picture element or color.
Artifact
Unwanted visual anomaly or defects generated by an input or output device, or by a software
operation, that degrade image quality.
Banding
An artifact of color gradation in computer imaging, when graduated colors break into larger blocks of
a single color, reducing the “smooth” look of a proper gradation.
Bit Depth
The number of colors a pixel can describe. 1-bit can describe two colors. 8-bit can describe 256. 24-bit
can describe 16.7 million colors.
Bitmap
The digital grid of pixels representing all the image data in a file.
Blend
In computer graphics software, the intermediate steps between two objects created when the
objects are merged together.
Brightness
The value of a pixel in a digital image, representing its lightness value from black to white. Usually
defined as brightness levels ranging in value from 0 (black) to 255 (white).
Calibration
The act of adjusting the color of one device relative to another, such as a monitor to a printer, or a
scanner to a film recorder. Or, it may be the process of adjusting the color of one device to some
established standard.
Corbis Technical Guidelines: Commercial and Editorial Photography - Page 16 of 22
Channel
A method for storing multiple layers of color information in a file. For instance, an RGB image would
have at least three channels. An alpha channel stores additional masking or color information.
Chroma
The color of an image element (pixel). Chroma is made up of saturation plus hue values but is separate
from the luminance value.
CIE LAB
Commission International del’Eclairage (pronounced "See-Lab"). A three-dimensional color model
developed by an international committee of color scientists (CIE). CIE LAB is a standard color model for
device independent color used in most color management systems. (L=Luminance, A=red to green, B=blue
to yellow.)
Clipping
The grouping (usually unwanted) of all tones or colors above or below a certain value into one composite
tone.
Cloning
The process of producing an exact copy of existing digital-image data, either to another image file or
copying an area of an image within the same file.
CMYK
Cyan, Magenta, Yellow, and Key (for Black). These colors make up the subtractive color space used in the
printing process.
Color Gamut
The range of colors that a particular device or process is able to reproduce or perceive. For example, the
visible gamut is much larger than that of offset printing because the human eye can perceive a much
wider range of colors that can be reproduced on a press.
Color Management System (CMS)
The process of calibration and profiling (describing) devices to ensure that input, monitor and output
devices closely match. A software and hardware system that is used to translate a file from one color
space to another utilizing profile information to obtain WYSIWYG results.
Color Profile
Also called device profile. This term refers to the relationship between the color models of the system
devices.
Color Saturation
Color Strength. A measure of color purity, or dilution, by a neutral.
Corbis Technical Guidelines: Commercial and Editorial Photography - Page 17 of 22
Color Space
Three-dimensional mathematical model enclosing all possible colors. The dimensions may be described in
various geometries giving rise to various spacings.
Colorimeter
A hardware device designed to measure color by responding to light in a method that is similar to the
human eye. This device is affected by surrounding light and is used primarily for monitor calibration.
Compression
The reduction of data to reduce file size for storage. Compression can be “lossy” (such as JPEG) or
“lossless” (such as TIFF LZW). Greater reduction is possible with lossy compression than with lossless
schemes.
Contrast
The difference between the lightest and darkest areas of an image. The greater the difference, the
greater the contrast. Conversely, an image with dark/dull highlights and light/gray shadows has little
contrast (flat). High contrast implies dark black and bright white content; Medium contrast implies a good
range from black to white; Low contrast implies a small range of values from black to white.
Delta E (DE)
A value representing the amount of change or difference between two colors within CIE LAB color space.
Industry research states that a difference of 6 DE or less is generally acceptable.
Density
The degree of opacity of an image.
Dot Gain
The optical and physical increase in dot size that occurs throughout the printing process from film to
finished product.
Dots Per Inch (DPI)
The resolution measurement for output from printers and imagesetters. An imagesetter may use a
resolution of 2400 dpi to create halftone screen film that is 150 lpi (lines per inch).
Encapsulated PostScript (EPS)
File format used to describe both bitmap and vector information.
File Compression
A technique for condensing a file so it takes up less space. Lossy compression deletes some data (JPEG
file format), while lossless compression retains all the original data (GIF/TIFF file format).
Corbis Technical Guidelines: Commercial and Editorial Photography - Page 18 of 22
Gamma
A measure of the range between highlight and shadow values. A gamma correction adjusts the contrast.
The higher the gamma, the higher the contrast.
Gradation
Transition between two colors or between black and white. Also known as a gradient.
Graphics Interchange Format (GIF)
File format for graphics with up to 256 colors and used primarily for the Web.
Gray Balance
The balance of colors combined to create gray or white without a noticeable color cast.
Grayscale
An image containing shades of gray as well as black and white.
Halo
A bright line tracing the edge of an image. This is usually an anomaly of excessive digital processing to
sharpen or compress an image.
Highlight
Refers to the lightest or brightest values in an image ranging from 0% to 25% (CMYK dot percentages) or
255 to 192 (RGB system values).
Histogram
A graphical display which represents the distribution of tones within an image. The horizontal coordinate
represents each pixel value possible from black to white. The vertical values indicate the number of
pixels in the image that occur at each value level.
HSB
Hue, Saturation, and Brightness. A color model that describes color by hue, saturation, and brightness.
Hue
A term used to describe the entire range of colors of the spectrum.
Interpolation
A technique for increasing the size of a graphic file by creating pixels.
Joint Photographic Experts Group (JPEG)
A lossy compression method, used in a file format referred to by the same name, popular for storing large
bitmap images and for displaying images on the Web.
Corbis Technical Guidelines: Commercial and Editorial Photography - Page 19 of 22
Kelvin (k)
Used to measure the color of light based on a scale created from the color changes that occur when a
black object is heated to different temperatures. Normal mid-day sunlight is considered 5000 Kelvin.
Lower temperature light (less than 5000 k) is more red or yellow, while higher temperature light is more
blue.
LAB Color Space (See definition for CIE LAB.)
Lossless Compression
Refers to a method of compression, such as LZW, that reduces the size of a file without any loss of data.
Lossy Compression
Refers to a method of compression, such as JPEG, that reduces the size of a file with varying or
adjustable loss of data.
Midtone
Refers to the middle range values in an image around 50% (CMYK) or 128 (RGB system value).
Moiré
An undesirable artifact or pattern that can appear in output film or scanned digital files.
Monitor Calibration
The process used to bring a monitor’s display of color, saturation and brightness into synchronization with
a particular standard, another monitor or a final output device.
Neutral
A tone that is created from a balance of all three primary colors such as browns, tans, whites or grays.
Open Prepress Interface (OPI)
A technique for using a low-resolution image as a placeholder for a high-resolution image that is
automatically swapped during output. See also APR.
Pixel
Short for Picture Element. The smallest unit of image data.
Pixelization
Jaggedness in an image that results when pixels have been added to the file (can be an artifact of
interpolation).
Pixels Per Inch (PPI)
The unit of measure used to describe the resolution of an image.
Corbis Technical Guidelines: Commercial and Editorial Photography - Page 20 of 22
Posterization
When you break down the tonal range of a given color into fewer differentiations, you begin to see the
breaks between the individual tones.
Profile
Also called characterization. After a device has been calibrated to a known standard, a profile can be
created to describe the device's ability to reproduce color. A profile is the mapping of a device's gamut in
the CIE LAB color space. In a color management workflow, a profile is used to accurately translate a file
from one device's gamut to another.
Proof – soft/hard, contract or content
Output designed to represent an image or file before it is printed. Soft proofs refer to proofs on the
monitor; hard proofs refer to proofs made on various substrates such as overlay or laminate proofs. Hard
proofs generally are either for content or contractual purposes. Contract proofs are high-quality, highcost proofs intended to represent the color quality of the final printed output accurately. Content proofs
are of lower cost and quality, and are intended only to indicate the content and position of the image or
page.
Resolution
A unit of measure. Image resolution measures the amount of data per inch in an image. Printer
resolution measures the amount of detail that a printer or imagesetter can create. Scanner resolution
measures the amount of detail that a scanner can capture.
RGB
Red, Green and Blue. These colors make up the additive color space on your monitor or your scanner.
Saturation
The percentage of white in a color. Zero percent saturation is full white. 100% saturation is no white.
Shadows
Refers to the darkest values in an image ranging from 75% to 100% dot (CMYK) or 64 to 0 (RGB system
values).
Spectral Highlight
An area in an image caused by the bright reflection on a shiny surface that has no detail, therefore can be
printed using little or no dot.
Spectrophotometer
A hardware device that is used to measure the characteristics of light based on where it resides within
the visible spectrum. Used primarily for fingerprinting colors and calibrating output devices.
Corbis Technical Guidelines: Commercial and Editorial Photography - Page 21 of 22
Standardization for Web Offset Publications (SWOP)
Ink, screen ruling and stock specifications established to maintain consistent, predictable results among
printed publications.
Subtractive Color Theory
The principle of reproducing color by reflected light. Light shines on a surface and is either reflected or
absorbed by pigments applied to the surface. Cyan subtracts red light, magenta subtracts green light, and
yellow subtracts blue light. The presence of all pigment theoretically creates black (all light absorbed),
and the absence of all pigment creates white (all light reflected).
Subtractive Primaries The three main colors used in the printing process: cyan, magenta, and yellow. In
theory, these are the only colors needed to create all the colors in the visible spectrum. Due to the
impurities of ink and variation in the reflective surface, black is used to assure true blacks and deeper
shadows.
Tagged Image File Format (TIFF)
A popular file format commonly used for describing bitmap image data.
Under Color Removal (UCR)
The process of creating the black separation by reducing the cyan, magenta and yellow inks from neutral
shadow areas and replacing them with black. By reducing these three colors, you use less ink and have
better shadow detail and more controllable neutrals.
Unsharp Masking (USM)
The process of creating the appearance of definition or sharpness in an image by increasing contrast along
the edges of areas where there are color or luminance changes. In Adobe Photoshop, you perform USM by
applying the various settings of Unsharp Masking Filter.
Vector
Drawn images, not photographic or bitmap, composed of a series of points connected with straight or
curved lines. These can be filled with color, gradients or patterns. These images are not subject to the
resolution or sizing issues of bitmap images.
Visible Spectrum
The portion of the electromagnetic spectrum that is visible to the human eye. Radio waves, ultraviolet,
infrared and gamma rays are portions invisible to the unaided eye.
WYSIWYG
What You See Is What You Get
Corbis Technical Guidelines: Commercial and Editorial Photography - Page 22 of 22