PreprintPDF Available

Technical Appendix: ReACH Declaration

Preprints and early-stage research may not have been peer reviewed yet.
Technical Appendix: ReACH Declaration
Collaborative version: 13 June 2018
Note on the authors: This first version of the ReACH technical guidelines was developed as
a collective project by the ReACH technical committee convened by the Victoria and Albert
Museum as part of the ReACH initiative and hosted by the Abu Dhabi Department for
Culture and Tourism in April 2018.
The following individuals contributed: Anais Aguerre, Chance Coughenour, Eugene Ch’ng,
Salma ElDardiry, Michael A. Keller, Paul Keller, Otto Lowe, Vladimir Opredelenov, Richard
Palmer, Daniel Pett, Merete Sanderhoff, Stefan Simon, Stuart K. Snydman, Daniel Suarez
and Rana Zureikat.
Technical Appendix: ReACH Declaration 1
The scope of this document 3
1. Creation of the derived work 4
1.1 Data capture 4
1.1.1 Methodology 4
1.2 Data processing 5
1.3 Production of high quality model(s) 5
2 The Digital Divide 6
3. Digital archiving and preservation 6
4. Sharing of the derived record and associated data 7
4.1 Publication of derivations 7
4.2 Access to derived and captured data 8
4.3 Licensing 9
4.4 Metadata and paradata 10
4.4.1 Metadata 10
4.4.2 Paradata 11
Reference materials 12
Standards and previous working groups 12
Online Guidance 13
Digital Preservation Platforms and Services 15
Technical definition of terms 16
Checklist for storage of assets 19
Meta and Para data 20
Metadata 20
Paradata 20
Example of image standards for documenting objects from the Pushkin Museum 22
Object dimensionality 22
Flat objects 22
Three-dimensional objects 22
B. Assessment of primary (basic, mandatory) processing 22
An additional assessment of images-results of panoramic (multi-frame) shooting 23
Additional color grading 24
The scope of this document
This technical appendix is a response to the publication of the ReACH Declaration in
December 2017. It was developed between April and June 2018. It is intended that this
document will be revised regularly and will remain technology agnostic and vendor neutral.
The ambition is twofold; firstly to clarify certain technical terms mentioned in the Declaration
and secondly to provide suggested guidance for the practical work of data capture (2D, 3D
and all forms of suitable recording of digital cultural heritage that emerge in the lifetime of
this guidance), storage and sharing to encourage a more harmonious practice and therefore
greater chance of collaborations and exchanges in the future.
The intent is that any guidance that is generated should be produced collaboratively with the
wider community that is currently engaged in any form of digital documentation using
imaging techniques and with reference to previous efforts such as the London Charter,
Principles of Seville, the European Commission’s 3D-ICONS research, CARARE and
ongoing work with CS3DP.
This dynamic document is divided into a technical guidance and various appendices and
1. Creation of the derived work
Several steps and interventions are acknowledged in the creation of a derived work
(“Record”) and those engaged in such endeavours are encouraged to follow article 14 - 16 of
the Declaration.
When creating reproductions/derived word (“Record”), due diligence must be applied to
determining whether any copyright, moral or religious right can be asserted over an object by
a living artist, relative, indigenous nation and whether there any cultural/ religious
sensitivities or contractual agreements that may prevent the reproduction being created with
the chosen method.
If this is the case, and it is possible to communicate with the affected rights holders or
communities, one should only proceed with such reproductions with the express permission
of the such rights holders or affected communities.
1.1 Data capture
The Declaration is technological agnostic, vendor neutral, and therefore the Steward of
Works (hereafter referred to as the Steward) should commission the appropriate method for
data capture based upon circumstances
, software and hardware availability, expertise and
budget. These methods include (but are not exclusive to) well documented processes (or the
Steward should endeavour to document their processes) such as conventional photography,
reflectance transformation imaging, photogrammetry, computed tomography, structured light
and laser scanning
1.1.1 Methodology
Stewards are encouraged to employ one of these standardised methodologies (with
reference to Article 2) that meets the constraints of the capturing institution (Article 14).
Many factors will influence the decision making when choosing the appropriate method for
The technique chosen is appropriate, non-invasive and maintains the integrity of
preservation of the subject’s condition;
Derived data can be hosted and preserved;
Technology can easily be accessed, documented and a plan for knowledge transfer
should be in place;
Circumstances for data capture of cultural heritage objects can vary greatly with the environment in
which the procedure can take place. For example, the location may be remote, without power and
have cultural sensitivity attached.
The list of data capture methods will expand and contract as new technologies come and go.
1.2 Data processing
Data should be processed using reproducible, attributable and ethical methods (for example
if crowdsourcing or other public labour is used, full attribution should be made). The onus
should be on Stewards to use software packages that are easily obtained and are
encouraged not to be digitally divisive (see section 2.) All steps should be documented
open and accessible formats, for a reproducible methodology to be enabled and for
knowledge transfer to occur (see section 3.4.2 below).
Data processing will involve steps, informed interventions and decisions made by both
machines and humans and vary depending on data capture methodology and hardware.
These steps might include:
Point cloud assembly - processing the image to derive measured coordinate based
points that can be used to assemble the shape of the object
Mesh cleaning - this can refer to smoothing, deletion, hole filling or various
techniques used to repair missing data or removing extraneous data.
Image manipulation - colour balancing and enhancing, image masking, resampling
and other methods to produce the best quality finish.
Image stitching - for example, documentation of large 2d works of art often required
manual or computerised joining together of images.
All data processing can introduce margins for error, so it is vital that paradata documents
methods employed, equipment and software used, who was involved and what decisions
were made to create the derivation.
All data captured, the RAW files in the least, should be stored securely and not discarded
unless they do not carry relevant information.
1.3 Production of high quality model(s)
The Steward of Works should endeavour to create high quality models to be generated from
the captured data set; this is of course dependent on the quality of the raw dataset and if
limitations are apparent, these should be documented within para/meta data. Models and
representations (of all types) should be made at the highest resolution that processing
equipment and source data will allow, and subsequent derived surrogates should be
processed and published. Models should be in interoperable formats (see definitions) that
can be converted as technology progresses.
It is not proposed to prescribe standards for what meets high quality due to the accelerate
pace of change that is applied to 3D technological development. Moreover, publication
online is governed by many factors, so derived lower resolution surrogates are required.
This document covers a multitude of data capture methods. The community is encouraged to openly
discuss and formulate documentation standards.
However, there are existing quality standards which have been developed and could serve
as useful guidelines: the FADGI quality standards and the Metamorphoze standards.
2 The Digital Divide
Usually, the Digital Divide is defined as:
“the gulf between those who have ready access to computers and the Internet, and
those who do not.”
In this case, many collaborating institutions, signatories and subscribers to the Declaration
may have little to no access to high end equipment and expertise and they should therefore
be furnished with accessible and affordable (and current) methods, and with access to
knowledge sharing opportunities and hardware sharing pools.
Signatories to the Declaration could contribute to a register of equipment held and details of
how they could be loaned (either with staff or without) for the mutual benefit of all. It should
be noted that software and hardware will have a defined life cycle and equipment
obsolescence is a high risk factor.
Signatories to the Declaration should be fully aware of the Digital Divide when
commissioning reproductions. Methods and techniques for data capture should be
documented and shared by Stewards and associates to ensure reproducibility of process.
3. Digital archiving and preservation
The Steward should preserve the derived data in an accessible form that will allow for
reproducible methodologies to be created and implemented as referenced in articles 6 - 9 of
the Declaration.
Digital Preservation of these data is acknowledged as a troublesome issue for Stewards, but
it must be adequately addressed and scoped when engaging in these methods. Different
methods of capture generate varying volumes of data
and subsequently Stewards should
be aware of the burden capture may place on their institution.
Therefore Stewards are strongly encouraged to investigate storage/archival and
preservation facilities, long term costs, frequency of access needed and length or retention
of these data before data capture begins. Stewards should define their preservation goals,
intentions and abilities, and consider the wide range of preservation solutions available. It is
essential that due regard is given to development of policy and practise that has little
disruptive influence on existing workflows. Extensive experience, methodologies, and
An example of higher end (affordable equipment) for file sizes would produce data such as RAW 20
- 80+ Megabytes per file, JPG at a specific resolution is at 5 - 20+ Megabytes per file. Therefore 100
RAW files at 20 Megabytes will produce 2 Gigabytes of data. A structured light scanner could produce
files in the Terabytes.
communities of practice in the community of research libraries offer insights into approaches
dealing with these challenges.
Data should be stored in facilities that allow for long term preservation of digital assets
storing on external hard drive or USB drive is not likely to ensure long term sustainability of
data. Stewards should look to digital archives in the University Sector or invest in their own
or shared digital preservation solution (public or private, as appropriate). In general,
Stewards should follow the ‘Lots of Copies Keeps Stuff Safe’ (LOCKSS) principles
. If these
principles are followed, data and derived records are more likely to survive long term.
4. Sharing of the derived record and associated data
4.1 Publication of derivations
Articles 10 to 13 of the ReACH Declaration champion the sharing of Record(s).
The Steward should be bound to make the Record(s) publicly available as soon as is
reasonably practicable using an appropriate method of publication or dissemination. Legal
requirements, rights of use, and sensitivity to art and cultural heritage of sentimental,
nationalistic, and religious value should be observed and clearly communicated alongside of
the published works.
The Steward should make best endeavours to ensure that the Record(s) are published in an
fashion, using appropriate non-proprietary/open and community standards
that will allow for manipulation and conversion as technology progresses. Care should be
taken when publishing works in corporate-owned platforms, terms of use should be read so
that rights are properly managed.
Uses of the final reproduction could include (but is not limited to) open source and hosted
online or in galleries/museums/interpretation centres for mobile and desktop 2D/3D viewers,
prints and tactile materials (including haptics), immersive environments (sound, light, visual),
Virtual Reality, Augmented Reality, Mixed Reality, games engines, interactive 2D/3D
systems, and 3D manufacturing.
All of these dissemination methods have the potential for reaching new audiences globally,
transmitted via the Internet
often at very low cost, therefore satisfying Article 11 of the
Cultural exploitation should be avoided as a first priority. However, there will be some groups
or individuals (i.e., marginalised groups, ethnic minority, etc.) who may make use of the
Backups will include tape drives, multiple computers, cloud providers, institutional servers.
In addition, these resources should be hosted and accessible via interoperable systems, based on
common APIs and open web standards for easy sharing, comparison, annotation and re-use.
This list is not exhaustive, new uses for derived works are being created quickly.
It should be noted that sometimes, a surrogate for the highest quality work may be used online to
enable usage due to size of files.
digital resources which Stewards, or they themselves have captured for creating a livelihood
for themselves or for innovation, and thus empowering the local economy. Stewards should
not prevent groups with direct ties to the original objects from exploiting their own cultural
resources (including derived representations) for their own (economic) benefit.
4.2 Access to derived and captured data
The Steward should make all data generated (where possible and not governed by
contractual or cultural reasons) during the recording process and subsequently derived data
(Article 6) available for (re)use. As technology improves, data may be reprocessed (as and
when required or through serendipitous reuse) and there is potential for higher quality works
to be produced. Data should be stored in an appropriate digital archive (institutional or
shared), with a commitment to long term digital preservation, and when practicable a Digital
Object Identifier (DOI/PURL) or other means should be assigned to the data compendia
which allows for citation (to complement the appropriate meta/para data).
Examples of captured and derived data that access and storage needs creating for include
(but are not exclusive):
1. Photographs - RAW (native formats), JPEG, TIFF, etc
2. Point clouds - OC3, etc
3. Models - OBJ, PLY, STL, etc
4. Image masks - PNG, JPEG, etc
5. Camera orientations - XML, etc
6. Metadata files - MD, XML, DOC, etc
It is suggested that Stewards follow the Findable, Accessible, Interoperable and
Reusable (FAIR) principles as laid out below (see
To be Findable:
F1. (meta)data are assigned a globally unique and persistent identifier
F2. data are described with rich metadata (defined by R1 below)
F3. metadata clearly and explicitly include the identifier of the data it describes
F4. (meta)data are registered or indexed in a searchable resource
To be Accessible:
A1. (meta)data are retrievable by their identifier using a standardized
communications protocol
A1.1 the protocol is open, free, and universally implementable
Commented [1]: @ Daniel - I am not sure it is so clear.
Commented [2]: I'll have a go. This was Eugene's
wording, which I think we agreed in AD.
Commented [3]: Thanks Anais, Daniel.
The idea is that whilst we are attempting to protect the
exploitation of particular objects captured from a culture
from being used by external groups without benefiting
them, we wouldn't want our technical policy to prevent
them from fully exploiting their own cultural resources,
and for their own economic benefits.
Commented [4]: so maybe make a declarative
statement at the end. Say something like "Stewards
should not prevent groups with direct ties to the original
objects from exploiting their own cultural ressources
(including derived works) for their own (economic)
Commented [5]: That works for me, incorporated.
Commented [6]: +1
A1.2 the protocol allows for an authentication and authorization procedure, where
A2. metadata are accessible, even when the data are no longer available
To be Interoperable:
I1. (meta)data use a formal, accessible, shared, and broadly applicable language for
knowledge representation.
I2. (meta)data use vocabularies that follow FAIR principles
I3. (meta)data include qualified references to other (meta)data
To be Reusable:
R1. meta(data) are richly described with a plurality of accurate and relevant attributes
R1.1. (meta)data are released with a clear and accessible data usage license
R1.2. (meta)data are associated with detailed provenance
R1.3. (meta)data meet domain-relevant community standards
4.3 Licensing
In order to comply with Article 10 of the ReACH Declaration, which aims to ensure the
availability of records to the public "for personal use and enjoyment and for non-commercial
research, educational, scientific and scholarly uses", Stewards should use open licenses
whenever possible. Machine readable open licenses and rights statements increase
the discoverability of published data and empower the public to use published
materials while respecting copyright. As such they are an important element of any
strategy aimed at sharing access to cultural resources.
When licensing published data (including metadata, captured data and derived data),
it is important that licenses are applied in accordance with any underlying copyrights.
It is both possible that Stewards may want to share data without having the rights to do so or
that Stewards may want to impose conditions on the sharing of data where they do not have
the right to do so. While it is beyond the scope of this technical appendix to provide case by
case advice there are a number of general guidelines for adhering with the sharing
objectives of the ReACH Declaration:
When publishing metadata and paradata, Stewards should waive any rights that
they may have in such data (most factual data will not be covered by copyright to
begin with) so that such data can be used by anyone without any restrictions. This
will increase discoverability. Within the cultural heritage community the Creative
Commons Zero Public Domain license has become the de-facto standard legal tool
for publishing metadata. We therefore encourage the use of CCO for metadata and
When sharing captured data, derived data or any other digital representation of
Works of Art and Cultural Heritage, the underlying rights status of the works
determines the conditions for sharing.
- When the underlying Works are free from rights (for example because copyright
has expired or has never existed) then the Steward should make these records
available with minimal conditions as possible.
- if no new rights have been been created in the process of reproducing
the work (in many countries faithful reproductions of a Work are not
protected by copyright or similar rights), then the Steward should label any
captured data, derived data or any other digital representation as free from
rights for example by applying the Creative Commons Public Domain mark.
- if new rights have been created in the process of reproducing the Work
(as it is the case in a number of countries and more generally when the
reproduction has an original character), then the Steward should either waive
these rights (by applying the above mentioned CC0 waiver) or license any
captured data, derived data or other digital representations of the Work under
an open licenses (such as one of the six Creative Commons licenses)
- When the underlying Work is not free from rights or is of a culturally sensitive
nature, reproductions (Records) should not be shared without consent from the
rights holders or affected communities. If they approve the sharing of any captured
data, derived data or other digital representations, then the conditions for sharing
should be approved by the rights holders/affected communities. In line with Article 10
of the ReACH Declaration, sharing under an open license (such as one of the six
Creative Commons licenses) is preferable. Should this not be acceptable, offers a number of machine readable rights statements that can
be used to convey more restrictive reuse conditions.
Regardless of the conditions under which Records are shared, Stewards should ensure that
these conditions are expressed in the associated metadata via standardised, machine
readable rights statements and/or licenses.
In accordance with Article 13 if the ReACH Declaration, Stewards should provide attribution
to the original author/creators of the Works and, where practicable, provide credit to those
involved in the process of documenting and producing Records of Works and should
encourage any subsequent users to do the same. Where captured data, derived data or
other digital representations of the Work are protected by copyright, attribution of the rights
holders can be legally required, in all other cases it should be encouraged.
4.4 Metadata and paradata
4.4.1 Metadata
The Steward should record, within their institution’s digital asset management systems (if
they exist), the appropriate metadata relating to the Work and the Record. Standards for this
are and have been in development now for a reasonable length of time (see for example
CIDOC and CIPA) and at some point an agreed metadata standard will exist (as per Article
16). Metadata should be comprehensive and provide a rich means via which Records can
Commented [7]: @ Daniel - please can we include a
link here?
Commented [8]: I think this is the correct one raised this, so probably best
he confirms.
be discovered, retrieved, manipulated and reused by humans and machines (in this context
we refer to Linked Data principles and protocols).
4.4.2 Paradata
The Steward should also record paradata relating to methods employed to generate the
derived work (Record), detailing how, why and what the reproduction was made and
required for. The recording of paradata are often overlooked in the heritage sector, but it is of
paramount importance to ensure future usage and interpretation. Paradata should make
reference to:
Agency - the choices made in relation to production of the representation
Authority - the institution or person creating the representation provides validity to
the reproduction
Transparency - the reasoning as to why the representation was made, the purposes
behind the processes
Authenticity - how closely the representation is to the Work. Has work been done to
the representation to enhance or obscure detail.
Reference materials
This section highlights some relevant material that the ReACH Technical Committee took
into consideration to develop the first edition of the ReACH Technical Guidelines conceived
as the technical appendix to the ReACH Declaration.
Standards and previous working groups
3D Icons Guidelines (2014)
[Accessed 27th March 2018]
A. D'Andrea and K. Fernie, "CARARE 2.0: A metadata schema for 3D cultural
objects," 2013 Digital Heritage International Congress (Digital Heritage), Marseille,
2013, pp. 137-143.
doi: 10.1109/DigitalHeritage.2013.6744745 [Accessed 13th June 2018]
CIDOC Conceptual Reference Model
CIPA standards
Community Standards for 3d preservation [Accessed
27th March 2018]
FADGI quality standards
Linked Art Data model
The London Charter Version 2.1 February 2009
[Accessed 27th March 2018]
Metamorphoze quality standards:
OAIS Reference Model (ISO 14721): The fundamental standard for digital
The British Library’s guidelines on Digital Preservation
The Seville Principles 2011 Final Draft
principles/seville-principles [Accessed 13th June 2018]
UNESCO 2003 Charter on the Preservation of the Digital Heritage1
UNESCO/PERSIST Guidelines for the selection of digital heritage for long term
Online Guidance
These guidance pages are not exhaustive and can be added to at any time as and when
resources are suggested.
Crawford, A. How to set up a successful photogrammetry project
[Accessed 27th March 2018]
Factum Foundation (n.d) 3D Scanning for Cultural Heritage Conservation
[Accessed 13th June 2018]
Historic England (2017) Photogrammetric Applications for Cultural Heritage
Historic England (2017) 3D Laser Scanning for Heritage
Historic Environment Scotland (2018) Applied Digital Documentation in the
Historic Environment Short Guide 13
Kaleta, R, Bevan, A. & Keinan-Schoonbaert, A., Bonnachi. C. (2014) Photographic
Strategy for 3D Models of Artefacts
Lievendag, N., The Beginners guide to 3D scanning and photogrammetry on a
Pier 9: Guide to Artec Scanners (2014)
Weinberg, M. (2016) New Whitepaper on 3D Scanning and (the Lack of)
Digital Preservation Platforms and Services
A wide array of preservation platforms exist, but many do not provide services that can cope
with all forms of imaging formats. The Steward should choose the most suitable platform or
service for the records and data that they produce. Platforms and organisations of note
The Archaeological Data Service
The Digital Archaeological Record (TDaR)
Stanford suite of services
Internet Archive
Stewards and signatories should also take note that personal archiving can also be effective:
Personal archiving
Technical definition of terms
Agency is the capacity of an individual or organisation to act within an environment.
Asset in this case can mean the Work that will be documented, or the digital files that are
created during the documentation process. In the ReACH Declaration, the digital file is
defined as “Record”: digital recording or reproduction of a Work and the data generated in
the process of faithfully capturing images and data of the Work so as to create a high quality
digital or physical reproduction of the Work.
Captured data means the image(s) or other information taken during the documentation
Computed tomography means radiography in which a three-dimensional image of a body
structure is constructed by computer from a series of plane cross-sectional images made
along an axis called also computed axial tomography, computerized axial tomography,
computerised tomography.
Cultural Institutions refers to (but is not exclusive to) museums, galleries, archaeological
sites and monuments, interpretative centres, universities, libraries and community places of
interests. Note that in the ReACH Declaration, Steward has a broader sense. It means, any
governmental or private entity that owns or possesses Works held for the benefit of the
public. The term Steward is intended to be broadly construed and includes, but is not limited
to, museums, sites, monuments, libraries, repositories, archives, places of worship, whether
governmental, sovereign, or private.
Creative Commons licenses are a range of public copyright licenses that have been
created to enable the free distribution of an otherwise copyrighted Work or Record. This type
of license gives people the right to share, use, and build upon a work under the clauses
Decimation means the process of removing polygons, from a 3D geometric representation.
The outcome of this is to reduce the size of the representation for processing.
Derived data means the byproduct of working with the imagery captured during the
documentation process. Images could be processed to remove background(s) or features,
which derives new information.
Derived Work means the product of the imaging process that is used to document or
recreate the original subject/work. In the ReACH Declaration, it is referred to a “Record
defined as a digital recording or reproduction of a Work and the data generated in the
process of faithfully capturing images and data of the Work so as to create a high quality
digital or physical reproduction of the Work.
Digital Divide refers to the disparity of access to technology, the Internet and equipment
between those who do, and those who do not.
Digital Object Identifier (DOI) means a unique alphanumeric identification string which has
been assigned by a registration agency. This string identifies the content and provides a
permanent and persistent URI for its location on the Internet.
Digital Preservation means the series strategies and actions taken to promote the
availability and usability of digital information over time. This concept is intended to prevent
loss, and ensure the durability of digital cultural heritage over time.
High quality means a level of quality sufficient to constitute a representation of a Work as
faithful as possible. Derived representations or Record should be produced at the highest
level that the Steward’s hardware and software will allow at the time of production with scope
for reuse of raw data if processing techniques change.
Interoperable means the ability for different systems (software) to exchange and make use
of information.
LIDAR is the controlled deployment of laser beams via a rangefinder, from which accurate
measurements are captured and interpreted.
Linked Open Data means a set of technical design principles for sharing machine-readable
interlinked data on the Web.
Machine readable means data in a form that a computer can process.
Metadata refers to data that has been generated to describe and give enhanced information
about other data.
Non-invasive in this context, assumes that any method for recording the object or
monument (‘Work’) should not in anyway damage it. The appropriate method will be used to
facilitate the recording with adequate documentation of risk to the subject.
Non-proprietary means the use of agreed standards that are in the public domain or are
widely licensed so that a wide array of applications and software options exist.
Open access means that the Works (as defined in the Declaration’s definitions point B) are
available to all, with little if any restriction for reuse.
Open Source means software which allows all citizens the chance to view, modify, enhance
the underlying source code.
Photogrammetry is the application of photography to obtain accurate measurements of
physical objects or the environment (Photo - picture, grammetry - measurement) and is used
as a means to create a 3D representation.
Photomask means to mark-up an image for the extraneous data, signifying to the
processing software that this area should be ignored when creating a 3D derived work.
Point cloud means a group points held within a coordinate system. In this case we are
referring to a 3D grid using XYZ planes.
RDF means Resource Description Framework.
Reflectance Transformation Imaging or Polynomial Texture Mapping is a technique of
imaging and interactively displaying objects under varying lighting conditions to reveal
surface phenomena.
Reproducible research in this case refers to subscribing by default to the practice of
making accessible alongside the high quality 3D model the raw data, the source code (if
created) and well documented tools and methods used to created the work.
Scanning means the process from which 3D data is derived following capture via a sensor.
Structured light scanning refers to the projection of light (either as a narrow band or as a
pattern) onto a 3 dimensional object from which accurate measurements can be obtained.
URI means "Uniform Resource Identifier." A URI identifies the name and location of a file or
resource in a uniform format.
Work as defined in the ReACH Declaration, means a work of art or other cultural item. The
term Work is intended to be broadly construed and includes, but is not limited to, works of art
in all media and eras, e.g., paintings, works on paper, sculpture, murals, antiquities,
monuments, architecture and architectural elements and archeological sites.
Checklist for storage of assets
Raw data should include:
1. Photographs (RAW, JPEG, TIFF)
2. Point clouds
3. Image masks
4. Meshes
5. Textures
6. Meta and para data files
Final representations and surrogates should include:
1. Interoperable formats - PLY, OBJ, STL, etc
2. Textures - JPG, PNG, TIFF
3. Metadata and paradata files
Data and derived records should be stored:
1. According to the Lots of Copies Keeps Stuff Safe principles.
2. Locally and if possible on an institutional network drive
3. Remotely on secure, backed up storage (tape, cloud, hard drive etc)
Meta and Para data
The following metadata fields are suggestions for bare minimum. Schema for recording 3D
data as Linked Open Data and relational databases are being discussed at various
conferences worldwide and through different funding streams.
Examples include the CARARE extension for 3D metadata. Ideally metadata would be
created via online editing forms and from automatic extraction.
An attempt should made to unify these and produce a standardised shema including
RDF and incorporation into CIDOC-CRM.
Date range
Original creator/artist/architect(s)
Geographical location of original
Material of original work
Data capture dates(s)
Date capture by whom
Model size
Scale applied
The following suggests basic paradata relating to the creation of the reproduction (‘Record’)
and are open to suggest and revision. Ideally, paradata would be created via online editing
forms, stored within a relational or graph database.
Imaging method employed
Imaging equipment used
Reasoning for method chosen
Processing software used
Processing software version
Processing settings used
Image masking applied (photogrammetry)
Mesh processing which could have involved deliberate smoothing, patching or
Number of chunks
Quantity of photos
Raw data capture size
Data processor(s)
Date(s) of processing
Nature of mesh cleaning applied
Texture processing and post production
Computer specification used
Rights statement
Storage location(s) for raw and derived data
Level of skill associated with creator
Use case for the reproduction
Costs for capture and processing
Example of image standards for documenting
objects from the Pushkin Museum
A. Object dimensionality
Standards will vary depending on the object dimensions.
Flat objects
Paintings, graphics, documents, photographs, flat numismatics, etc.
1. The size of the object is less than A7 (7.4 × 10.5 cm): not less than 1200 dpi,
not less than 3500 × 4900 pixels for A7.
2. The size of the object from A7 (7.4 × 10.5 cm) to A6 (10.5 × 14.8 cm): not
less than 600 dpi, not less than 1700 × 2400 pixels for A7 and 2400 × 3500
pixels for A6.
3. The size of the object from A5 (14.8 × 21 cm) to A4 (21 × 29.7 cm): not less
than 400 dpi, not less than 2300 × 3300 pixels for A5 and 3300 × 4600 pixels
for A4.
4. The size of the object from A3 (29.7 × 42 cm) to A2 (42 × 59.4 cm): not less
than 300 dpi, not less than 3500 × 4900 pixels for A3 and 4900 × 7000 pixels
for A2.
5. The size of the object to A1 (59.4 × 84.1 cm): not less than 200 dpi, not less
than 4600 × 6600 pixels for A1.
6. The size of the object is more than A0 (84.1 × 118.9 cm): not less than 5700
× 8600 pixels.
Three-dimensional objects
Sculpture, DPI, archeology, volume numismatics, etc.
1. Regardless of the size, at least 5400 x 3600 pixels.
B. Assessment of primary (basic, mandatory) processing
a. Primary (basic, mandatory) processing of digital images of museum items includes:
1. Rotate the frame 90/180 degrees, align the horizon line / axis of the frame to
correctly display the subject;
2. Cropping;
3. White Balance Correction
4. Converting to TIF / TIFF format (Adobe RGB 1998) 8 or 16 bit without
compression / with LZW compression.
5. Correction of the exposure and sharpness of the image is made if necessary.
b. Digital images that have special purpose and converted to other formats (PSD, PNG,
etc.), color spaces (CMYK, Lab, Grayscale, etc.) that have layers, additional channels or
other differences, should have a file with a general view to this object in TIF / TIFF format
(Adobe RGB 1998) 8 or 16 bit without compression / with LZW compression, reduced in one
layer into 3 RGB channels.
c. Museum items should be displayed correctly relative to their upper and lower boundaries,
be located front (or in the angle that is required) without inclines and deformations distorting
its original boundaries and surface.
d. The museum item should be placed in the center of the frame with the minimum amount
of space around in all cases, except for artistic photography for special projects (printing
products, multimedia content, etc.), where an artistic background and creative framing is
e. The digital image must pass the white balance correction manually or automatically in
order to match the color gamut of the image to the color gamut of the original object. The
absence of white balance correction is allowed when streaming large volumes (from 70
subjects per day), but with the presence of a color target in the frame or a separate
f. Exposure compensation (exposure compensation) is allowed to compensate for explicit
exposure metering errors when shooting when the subject in the frame is significantly darker
/ lighter than the original.
g. Correction of sharpness (most often, sharpening) of a digital image during processing is
allowed, if it does not lead to the appearance of an obvious digital noise.
h. Do not allow the presence of glare on the object, interfering with the perception of its
volume, texture of the surface and other useful visual information.
An additional assessment of images-results of panoramic
(multi-frame) shooting
a. If the digital image was compiled from several fragments manually or programmatically
(that is, a panoramic / multi-frame shooting of the object was performed), it is necessary to
check the correctness of the connection of frames to the subject:
i. Presence of physical shifts in the frame boundaries, inconsistencies in lines,
textures, the presence of duplicate elements (doublet) or the absence of parts of the
image in the junctions;
ii. Presence of the staff, differing sharpness, white balance, exposure on one digital
b. If an incorrect connection of frames is detected, it is necessary to refer to the original
survey files (RAW) for evaluating the possibility of re-processing (in the case of errors at the
stage of connecting frames) or the need to re-shoot museum items.
Additional color grading
a. If necessary, as well as in cases of special high-precision artistic photography for special
projects (printing products, multimedia content, etc.), additional control over the correctness
of color reproduction using color targets, special color profiles for cameras or other means
(hardware and software) color accuracy.
b. Such checks should be carried out on special professional calibrated monitors having high
color accuracy. When digitally printing samples to control the color reproduction of files, the
printers must also be calibrated.
ResearchGate has not been able to resolve any citations for this publication.
ResearchGate has not been able to resolve any references for this publication.