Photo Metadata User Guide - ToDo Reference (2024)

On the left, the diagram shows a metadata input panel for IPTC fields in your software, only a few fields are shown here.

In the centre, the diagram shows how your software handles the metadata. The software creates a temporary internal structure to support the data in both IIM (in blue) and XMP (in red) formats.

You can see from the orange arrows that some field values are transferred to both the IIM and the XMP structures (Creator and Description) and others only to the XMP structure (Persons Shown and Digital Source Type). Fields added to the IPTC Photo Metadata specification’s Extension Schema (after 2005) exist only in XMP.

As some fields are replicated in IIM and XMP, the software has to keep these values in synch (dotted lines). This is less an issue for data transferred from the input panel than for data read from an image file.

On the right you see the structure of an image file with sections for metadata (headers) and a section for image data (the pixels). The diagram shows that the IIM and XMP data structures are stored in two different headers. In a JPEG file the headers are of type APP1 and APP13 and they may have additional internal identifiers for the metadata formats.

The data structures are embedded into the headers using different technical formats:

The green arrows show how data from the internal software structure is embedded in the headers of the image file. If this file is copied to another computer, then exactly this set of embedded bytes takes the metadata to the new location.

Data embedded in the image file can be read by software, so the process also works from right to left on the diagram, as shown by the bi-directional arrows. The software reads from the metadata headers and stores the information in its internal structures. The software has to verify that the IIM and XMP values are the same; if they are not, there are rules about which value should take precedence. At the end of this process IIM and XMP values are the same.

The values are then displayed on the panel on your computer screen and are ready for editing. Pressing Save, sends the data back to the image file for embedding as before.

Enter keywords to describe the visible and abstract content of the photograph. Keywords are in free text form, and may be single or compound terms.

Keywords are descriptive words added to an image to enable search and retrieval. They describe what is visible in the image and concepts associated with the image. Keywords are expressed as a list of terms. Keywords can be single or compound terms.

Values from the controlled vocabulary IPTC Subject Codes should be placed into the “Subject Code” field.

This field can be used to specify and categorise the content of a photograph by using one or more subjects as listed in the IPTC “Subject NewsCodes” taxonomy (available from http://cv.iptc.org/newscodes/subjectcode). Each subject term is represented as a code of 8 digits in an unordered list. Only subjects from this controlled vocabulary should be used in this field, free text keywords should be entered into the Keyword field.

This field structure is a generic way to add one or more terms, themes or named entities to describe the image.

Multiple terms may be used; each term must be taken from an identified Controlled Vocabulary. Terms may be from different Controlled Vocabularies.

This CV field enables users to enter terms about the image from specific controlled vocabularies. Terms from one or more vocabularies may be entered.

The structure is:

Globally unique identifier for controlled terms to describe the genre of the photo. The IPTC Genre vocabulary may be used http://cv.iptc.org/newscodes/genre or other genre vocabularies more focused on photography.

This field structure is a generic way to describe the genre of the photo with a value from any Controlled Vocabulary. (The Intellectual Genre actively supports only the use of an IPTC vocabulary.)

Multiple genre terms may be used; each term must be taken from an identified Controlled Vocabulary.

Genre Terms from one or more vocabularies may be entered.

The structure is:

This field is used to describe the scene of a photo using one or more terms from the IPTC "Scene-NewsCodes". You should only enter values from the IPTC Scene controlled vocabulary (available from http://www.newscodes.org and http://cv.iptc.org/newscodes/scene). Each IPTC Scene term is represented as a 6 digit numerical string in an unordered list.

Many professional photo applications have had a image rating feature for some time. These are typically shown as star ratings within a collection and are used to indicate the quality of an image; typically giving one star for entry-level photos, and reserving the higher numbered values for more special or unique images. Assigning a star rating as part of a workflow will make it easier to quickly find, sort, or filter out more valuable images from a grouping at a later point in time.

Photographers may use a method where any ‘keepers' from an assignment are given one star during an initial review. On a second pass they may give a two-star rating to those images deemed superior, or even three stars for those that are outstanding. These values may differ from what an agency or distributor uses, so they may be overwritten or re-evaluated. Some editors recommend that you think of this as a pyramid, with a 10 to 1 ratio between each level. This method will ensure you won’t end up with too many ‘special' photos in a collection.

To make sure you consistently apply the same image rating criteria each time, write down your rationale. Then put this text somewhere you can refer to each time you are editing.

Here is one photographer’s image rating rationale as an example:

Note that the star rating is done by the user/supplier and there is no universal standard for the rating between systems/collections.

A headline is a brief synopsis or summary of the contents of the photograph. Like a news story, the Headline should grab attention, and telegraph the content of the image to the audience. Headlines need to be succinct. Leave the supporting narrative for the Description field. Do not, however, confuse the Headline term with the Title term.

The Description field, often referred to as a ‘caption' is used to describe the who, what (and possibly where and when) and why of what is happening in the photograph. It can include people’s names, their role in the action, the location. Geographic location details should also be entered in the Location fields. The amount of detail included will depend on the image and whether the image is documentary or conceptual. Typically, editorial images come with complete caption text, while advertising images may not.

Some text about Alt Text and Extended Description.

Use this field to note the name of a person or persons shown in the image. Typically these would be recorded as they would be typed in a query, first name / last name (given name / surname).

Use this field structure to record details about each relevant and recognisable person(s) shown in the image. This might include links to a global online resource which lists the person uniquely with an identifier. There are fields to record physical characteristics and other details to help distinguish this person from others in the image.

These details are useful for identifying and distinguishing this person from others in the image.

The Additional Model Information field can be used to record information about the ethnicity and other facets of the person(s) (“model(s)”) appearing in the image. Use the Model Age field to note the age of model(s).

Age of the human model(s) at the time this image was taken in a model released image. If there is more than one model in the image, the ages can be listed in any order.

The user should be aware of any legal implications of providing ages for young models.

Age of the youngest model pictured in the image, at the time that the image was made. The user should be aware of any legal implications of providing ages for young models.

This field summarises the availability and scope of model releases authorising usage of the likenesses of persons appearing in the photograph.

There are four possible values:

We recommend that the PLUS controlled value Unlimited Model Releases (MR-UMR) be used sparingly, and encourage you to check the wording of the model release thoroughly before choosing this value.

Use this field for the ID of each available Model Release document. Be sure to give a unique number or name to all releases (both model and property), and record that information in this field. If you don’t already include an ID name/number on your releases, consider adding one as this will make it easier to cross reference.

The legacy Location fields - in most cases shown as sequence of stand-alone fields - are widely understood to express the location shown in the image. They can be used where it is important to display the location values in software which does not read Location Created and Location Shown field structures. Some software applications copy data from the Location fields to the field structure ‘Location Shown.'

The location where the image was created.

Use this field structure to specifically record the location where the photo was taken. If the location shown in the image is different from the location where the photo was taken then the IPTC field structure ‘Location Shown in the Image' should be used to note the difference. For example, if you are photographing a mountain with a telephoto lens from a distance, you may be standing on the other side of a state or even country border.

This field structure describes the location shown in the image. Where the subject of the image is in a different location to the camera the values should differ from those in ‘Location Created'.

Featured organisations can be described by name and code:

The Event field describes a specific named event associated with the image, e.g. Archimedes press conference, The Great Steamboat Race, Maui Classical Music Festival. Sub events of larger events can be included as in: XXXI Olympic Summer Games (Rio): opening ceremony.

TODO: add the Event Identifier

The Product Shown field structure is used to describe one to many products depicted by the image. The name of the product and a textual description can be applied to the corresponding fields. To identify the product a 14 digit GTIN (Global Trade Item Number) of the product should be applied to the GTIN field, GTIN-8 to GTIN-14 codes can be used.

This field structure is used to record information about artworks or other objects in the image, and includes descriptive, administrative and rights information. This category covers paintings, sculptures, objects, and other items of interest for cultural heritage such as archaeological finds.

Name of the creator of the image. Where the artist cannot or should not be identified, the name of a company or organisation may be use.

This property can be used to indicate the creator or creators of the image by name and identifier.

The job title of the person who created the photograph. For examples this might include titles such as: Staff Photographer, Independent Commercial Photographer, or staff writer. Since this is a qualifier for the Creator field, the Creator field must also be filled out.

The Contact Info fields provide a generic structure for storing contact information for the person or organisation that created this image.

Some text about this property/field.

The Copyright Notice contains information required to assert copyright in the image and should contain the name of the current copyright holder, whether an individual or a company. The format will differ according to the relevant copyright legislation. It may include the copyright symbol ©, the year of publication, and other commonly used terms such as ‘All Rights Reserved.' If an image is Public Domain, it can be indicated here.

For legal advice on asserting copyright, you should consult a lawyer.

Notes on usage rights (how the image may be used) should be provided in the “Rights Usage Terms” field.

Indicate the owner or owners of the copyright in the image, using name and identifier. Note that Copyright Owner, Image Creator, Image Source and Licensor may be the same or different entities.

The Credit Line shows how the image should be credited when published, as specified by the supplier of the image. The format varies for different suppliers and may contain: Agency Name, Photographer Name, Rights assertions. E.g. Agency/Photographer; © Photographer; Museum/Artist The Credit Line may contain information also listed in other fields such as Creator, Copyright Notice, Supplier.

The Source field is used to name parties with a role in the supply chain, such as agencies, originating organisations, or photographers. The Source field is useful for syndication where the original supplier agency or photographer is different from the end supplier.

The Web Statement of Rights can be used to link the viewer to a web page (by a URL) which provides a statement of the copyright ownership and usage rights of the image. In the Adobe ‘File Info' panel this field is called the ‘Copyright Info URL.'

This field summarises the availability and scope of property releases for the photograph.

There are four possible values:

Use this field to indicate the ID for each Property Release document. Ensure all releases (both model and property) are assigned a unique number, and record that information in this field.

This field is for free-text instructions on how the image may be legally used. E.g. ‘Permission is required from (Supplier or Creator) to publish this image' or ‘Licensed to (Customer) for use in (publication) until (date)'.

For more detailed licensing terms, you may use the PLUS ‘Media Selector', or another standardised vocabulary. This field may also be used to indicate a Creative Commons Licence assigned to the image.

This field structure identifies the most recent supplier of the image. This may be the copyright owner, creator, or another party in the supply chain, such as an agency or other distributor. This field structure may also be used for parties with a role known as provider.

The ID assigned to the image by the Image Supplier. Not to be confused with the Image Supplier ID, which identifies the supplier, not the image!

This field structure holds contact details for the person or entity authorised to licence the image. It includes Name, Identifier, Phone number, Fax Number, Email address, Web address. Up to 3 licensors may be entered.

A machine readable rights expression may include all or some of the terms and conditions of a licensing agreement. It communicates key information such as permissions, constraints and duties to allow for informed decisions as to how, where and when an image may be distributed to end users.

The Rights Expression may be included in two different ways:

This structure holds encoded rights expressions. The values are created by software outside the panel, using standardised rights expression languages such as MPEG 21, ODRL or RightsML.

This structure holds details of encoded rights expressions referenced by a link.

This field records the date and optionally the time the image was created. This can be derived from the Exif DateTimeOriginal if that is supported by the software.

More about dates and times and different software

The name of the person creating or editing the description of the image.

A short human readable reference for the image. It can be a text reference or a numeric reference, and serves primarily as an identifier. It has been used by photographers for their image filename, though since about 2008 IPTC now provides specific fields for image IDs like Digital Image GUID or Registry Entry (those wishing to, can use the Registry Entry. The Title field should not be confused with the Headline field which is a short descriptive field about the content of an image.

A short human readable reference for the image. It can be a text reference or a numeric reference, and serves primarily as an identifier. The Title field has often been used by photographers for the image filename, but IPTC now provides specific fields for image IDs including the Supplier’s Image ID, The Digital Image GUID, and the Registry Entry fields. The Title field should not be confused with the Headline field which is a short descriptive field about the content of an image, or with the AO Title field which contains the title of the artwork or object in the the image.

A number or textual identifier for the job for which the image was supplied. This field can allow job information to be tracked through the workflow.

A free text field for instructions to the receiver from the creator or supplier of the image. Instructions can may include details of embargoes, restrictions, or any other rights or technical information needed for the end use. Be aware that there are more specific rights expressions fields (see Rights Information and Licensing sections) which can be used.

A field structure used to describe a registry entry for the image. The record must include identifiers for the registry and the registered item as below:

These fields together define the maximum image size in pixel dimensions available from the original image (which may have been downsized).

This field indicates the media source from which the digital image was created. The values are taken from a controlled list, available at http://cv.iptc.org/newscodes/digitalsourcetype. See Guidance for using Digital Source Type in this document for more details.

A globally unique identifier (GUID) for the digital image. The identifier, may be created by technical equipment such as camera or scanner as early as possible in the workflow. The creation of the identifier must comply with the technical requirements for a GUID, and should ideally identify the equipment used. Once entered, the GUID should not be changed.

You can use the IPTC Image Region to record details for designated areas within a still photo using rectangles, circles and polygons. You can give each image region a name and an identifier (if desired) and note what type of role (see IPTC’s CV) the region plays and the type of content (see IPTC’s CV) within that region. (The use of both IPTC CVs is recommended but not mandatory.)

There are many times when you need to identify people within an image. This can be difficult especially when all the people don’t line up in nice ‘left-to-right’ rows. The Image Region feature gives you a way to: isolate each person’s face or body in an image (using a rectangle, circle or polygon), give it an identifier and name, indicate that the marked area is a subject area and indicate that the type of content is a human. Finally add the IPTC property Person Shown in Image to the Image Region with the name of the framed person.

If you create photo composites, the IPTC Image Region can be used to identify each of the different entities making up the composite and tie the provider’s name or copyright notice back to each.

Suggestions for how the image could be cropped to accommodate different layouts can now be embedded into the image itself. For example, you could mark a horizontal rectangle within a vertical image and indicate that the role for this marked area is a “landscape format cropping” for that image.

It should be possible to automatically map the face-tagging features in some cameras to the IPTC Image Regions so you would only have to add the name of the person to the regions. In addition, auto-tagging or image recognition systems could create image regions and auto-fill the embedded Image Region metadata fields.

This is a new feature, so it’s quite likely that there are other use cases which haven’t even been thought of yet. Talk to your software developers and ask them to implement this feature, and share with them your ideas for how you intend to use the IPTC Image Regions.

Read on if you are planning to implement Image Regions into your software, or are simply curious about what needs to happen under the hood to make the IPTC Image Regions feature work properly.

Metadata for one to many image regions can be embedded in the image files. In time, this data should be read automatically and could be transformed into data displaying the shapes of the regions within an HTML page or in special software. These image region boundaries could be shown in a layer over or above the image and should be identified by the color of the boundary or by an identifier shown next to the boundary. Additional details about each image region should be shown in the same page/view—either outside the image (with it identified as a reference) or when hovering the mouse over an image region.

If images are cropped or resized and if the coordinates of the boundary of an Image Region are not adjusted it is very likely that they no longer frame the intended region. Therefore, if an Image Region asserts to be about a male person and the boundary touches or encompasses a female person one can assume that the image region is no longer valid. The CIPA Standard DC-008 Exif metadata tag ‘SubjectArea’ can be mapped to an IPTC Image Region (see the Note about that), in this case the data of the Exif SubjectArea must also be adjusted.

As IPTC Image Regions employ and rely upon coordinates expressed by relative size values or pixel count, Image Region metadata is vulnerable to corruption (rendering the metadata inaccurate) should any of the following changes occur:

In addition, if Image Region metadata has been applied to a composite image (an image made up of two or more images), then Image Region metadata is vulnerable to corruption (rendering the metadata inaccurate) should any of the following changes occur to the various elements:

The IPTC invites and encourages developers to create solutions designed to allow IPTC Image Regions to survive image alterations. At the time of this feature release no such solutions are available. In the interim the IPTC recommends that users exercise caution in relying upon the IPTC Image Regions to identify or express metadata regarding people, objects or other subject matter appearing in a photograph, as this data may be inaccurate.

In particular, to mitigate legal liability, IPTC recommends that users exercise extreme caution if/when using the IPTC Image Regions to express rights-related information pertaining to any element/s of a photograph (such as copyright, property rights or model release information).

Headline and language below must be changed

Google displays three IPTC photo metadata fields, wherever available, for an image shown as search result. This tells the viewer who is the creator and who is the copyright holder of the image and what credit line should be shown next to the image. This information is taken from the IPTC photo metadata embedded in the image file.

It’s important to understand that IPTC Photo Metadata is actually embedded in the image binary file. You can’t add HTML tags or schema.org markup to add this metadata. But never fear - there are some tools you can use to edit the fields.

We maintain a list of tools for editing IPTC Photo Metadata. Here are a few of the major tools we cover there:

Each of these tools will allow you to edit fields a slightly different way. Usually there is some kind of “properties panel” or “metadata window” that lets you view and edit all embedded metadata fields.

Your site’s digital asset management system, content management system, image management system or content delivery network may be stripping out embedded metadata fields. Some systems do this with the best of intentions, thinking that it will save a few bytes of bandwidth, but stripping out metadata actually infringes on the copyright holders' rights and may even be illegal in some countries.

You should use a DAM and CMS that respects and conserves IPTC and XMP embedded metadata, and ensure that any configuration options that strip out metadata are turned off. Also you may need to look at image cropping and manipulation plugins for your CMS - for example the ImageMagick WordPress library retains embedded metadata, but some others strip it out.