Fwd: [lib.dig.pres] Archival formats for 3D objects?

[Susan Borda]

Hi CS3DP-

Do you have any advice for Scott?

Thanks,

susan

---------- Forwarded message ---------
From: Scott Prater <lib.di...@lists.btaa.org>
Date: Mon, Jun 24, 2024 at 1:43 PM
Subject: [lib.dig.pres] Archival formats for 3D objects?
To: lib.di...@lists.btaa.org <lib.di...@lists.btaa.org>
Cc: Steven Dast <steve...@wisc.edu>, David Lee <davi...@wisc.edu>

Hello, everyone --

We have some 3D scanned objects incoming into our digital collections, whose files we can receive in either Wavefront (.obj) or Stereolithography (.stl) format, or both. The LoC preservation recommendations deem either of them as acceptable formats (there is no preferred format): https://www.loc.gov/preservation/resources/rfs/design3D.html#scanned-3d Both are plain-text formats.

Has anyone here worked with these 3D scan formats, either in an archival or display context?  Do you have a preference for one format over another, or any warnings or recommendations regarding them?  We did a few sample exports, and the sizes aren't terrible (~50 - 100MB per object), and looking at the sustainability factors, they both seem okay, not great – there seems to be more widespread application support for .stl files, though it seems to be less information-rich than .obj files.  On the other hand, there is no current published specification for .stl files, while there is a published spec for .obj.  Right now, I'm leaning towards the Wavefront .obj format for preservation purposes, and considering storing the .stl file, also, just in case.

thanks in advance,

-- Scott

--

Scott Prater

Digital Library Architect

UW Digital Collections Center

University of Wisconsin - Madison

 

--

Susan Borda

Digital Preservation Projects Manager

Digital Preservation Unit

University of Michigan Libraries

Buhr Building

sbo...@umich.edu

My office phone number is temporarily disconnected while I work remotely due to COVID-19. Please contact me via email.

 

[Vera Moitinho]

Hi Susan,

Very briefly, perhaps something Scott would like to be aware of:

The STL file format describes the 3D geometry, but cannot include image/colored texture (no colored decorations, paintings, spots, etc.). It is often used for 3D printing too.

Whereas the OBJ format describes both 3D geometry and image/coloured texture, in separate linked files though.

Cheers,

Vera

--

Vera Moitinho de Almeida, PhD

Coordinator, Senior Researcher

CODA - Centre for Digital Culture and Innovation

Faculty of Arts and Humanities | University of Porto | Portugal

 

Virus-free.www.avg.com

--
You received this message because you are subscribed to the Google Groups "Community Standards for 3D Data (CS3DP)" group.
To unsubscribe from this group and stop receiving emails from it, send an email to community-standards-for-3d-data...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/community-standards-for-3d-data-preservation-cs3dp/CAKicaywG%3DYC%3DE4TESA8Op_kWkE8nC1vcKKEYSCng-zt71e-Xmw%40mail.gmail.com.

[Mark Hellar]

If you are looking for a file format suitable for 3D printing that can contain color metadata, have a look at 3mf https://hawkridgesys.com/blog/3mf-the-file-format-for-the-future-of-3d-printing

To view this discussion on the web visit https://groups.google.com/d/msgid/community-standards-for-3d-data-preservation-cs3dp/CABS-%2BuVFR3REiM4rK2a-Sn%3DkGGfNQgAvParoiZPrrcYEAiTQdA%40mail.gmail.com.

[Ethan Gruber]

There's a lot of software support for both opening Wavefront obj in 3D modeling software and viewing the models in a wide variety of JavaScript libraries. For archival purposes, I think it's important to choose a format that is text based and not binary, if you are considering reuse and migration of files 20+ years down the road. I think PLY is also a viable archival format that has a lot of software support.

Both can point to texture files, so these are what I'd recommend for architectural models with textures and models generated by photogrammetry. These are probably the most common cultural heritage use cases, but maybe there's more demand for STL from science and engineering: laser scans, CT/MRI scans, etc.

Ethan

To view this discussion on the web visit https://groups.google.com/d/msgid/community-standards-for-3d-data-preservation-cs3dp/CA%2Byj544YwxAqBGvAn3TF9yP9a_PrTrf97rQ0DazFF%2BNBJjgYHg%40mail.gmail.com.

[Deidre Brin]

The ADS Guides to Good Practice section on file formats for 3D models might be helpful here, even if you're not archiving archaeological models. 

https://archaeologydataservice.ac.uk/help-guidance/guides-to-good-practice/data-analysis-and-visualisation/3d-models/creating-3d-data/file-formats/

Best,

Deidre

--

Deidre Brin (dee-dree)

Digital Archaeology Lab and Data Publication Director

Cotsen Institute of Archaeology

308 Charles E. Young Drive North

Box 951519, A322 Fowler

Los Angeles, Ca 90095

www.dal.ucla.edu

The Cotsen Institute of Archaeology at UCLA acknowledges the Gabrielino/Tongva peoples as the traditional land caretakers of Tovaangar (the Los Angeles basin and So. Channel Islands). As a land grant institution, we pay our respects to the Honuukvetam (Ancestors), ‘Ahiihirom (Elders), and ‘Eyoohiinkem (our relatives/relations) past, present, and emerging.

To view this discussion on the web visit https://groups.google.com/d/msgid/community-standards-for-3d-data-preservation-cs3dp/CAGm-SsVLvxFNTc-LPYK27kj-tNYnDSShA1%3Dg7jvuBKgf01ATGg%40mail.gmail.com.

[Kieron Niven]

Thanks for pointing to that, Deidre!

Just to add, we've been working with OBJ as a preferred format for a fair amount of time now with no real issues. As Vera pointed out, there are subtle differences between OBJ, PLY, and STL capabilities, and the OBJ format can be extended depending on the software. In a case like this I'd go with the depositor's preferred format if they're equally acceptable but, for preservation, obviously go for the plain text option over the binary where possible. The most common issue we come across is the broken link between the OBJ and related texture files (MTL and image e.g. JPG) where files have been renamed through another process, something to watch out for.

Cheers,

Kieron

Kieron Niven orcid.org/0000-0002-0537-9238  Digital Archivist: Data Standards
	Archaeology Data Service Department of Archaeology, University of York, The King’s Manor, York, YO1 7EP Email Disclaimer  |  Privacy Policy

To view this discussion on the web visit https://groups.google.com/d/msgid/community-standards-for-3d-data-preservation-cs3dp/CAJTyFDUK%3DLF3XJP0sGfja0JT-2BSfhLeDwEBerM%2Bw5E-biBxTg%40mail.gmail.com.

[Polys, Nicholas]

Hi all~

thanks for the discussion

I wanted to share a few data points that have not come up yet...

We have been digitizing our insect collection. 

We went through many of these tradeoffs folks here have mentioned: 

• dead formats
• no spec or standards
• lossy data (no metadata)
• lack of conformance in software tools

I will propose another perspective:

• start from the information requirements you want to represent
• use International Standards

For several reasons mentioned above, we prefer Extensible 3D (X3D) as an archival format.

It can be validated. It is and ISO/IEC standard. It can hold metadata as well as geospatial, volumetric, and interaction and animation information. ...

Being a Standard, carrying color, and being 3d printable... being extensible if you need a little or a lot of 3D capabilities...  this is why the NIH uses X3D as the enterprise backbone and deliver for their 3dprint.nih.gov.

Also having a process and community around the technology has made it durable and adaptable: 3D models from VRML (1998-present) and X3D still run and faster than ever across modern Caves, mobiles, headsets !

You can learn more about X3D here:

• https://webx3d.org/ and https://www.web3d.org/
• https://www.loc.gov/preservation/digital/formats/fdd/fdd000490.shtml
• 29 years of peer reviewed papers in ACM DL: https://dl.acm.org/action/doSearch?ConceptID=120550&sortBy=EpubDate_desc&startPage=

You can read a more scholarly treatment about the 3D insect project here:

Wen Nie Ng, Alex Kinnaman, and Nathan Hall. 2022. Levels of Representation and Data Infrastructures in Entomo-3D: An applied research approach to addressing metadata curation issues to support preservation and access of 3D data. In Proceedings of the 27th International Conference on 3D Web Technology (Web3D '22). Association for Computing Machinery, New York, NY, USA, Article 6, 1–6. https://doi.org/10.1145/3564533.3564573

and

Kinnaman A, Saverot M, Chen Y, Long K, Polys NF, Nesbit S, Stocker M, Xiao S, Hall N. 3D Preservation Models and Modalities: Advancing Research Reproducibility and Capacity at Virginia Tech. IniPRES 2021.

Generally we use scripting, metashape, blender, meshlab in production and then publish to the Web with X3DOM or X_ITE.

I am happy to continue the discussion and create a community of practice here!

Feel free to reach out if you have further questions!

with best regards,

 n_polys

--

Nicholas F. Polys, Ph.D.

Director of Visual Computing
Virginia Tech Research Computing

Affiliate Professor
Virginia Tech Department of Computer Science

https://people.cs.vt.edu/~npolys/ 

---

From: 'Kieron Niven' via Community Standards for 3D Data (CS3DP) <community-standards-for-3...@googlegroups.com>
Sent: Tuesday, June 25, 2024 4:42 AM
To: community-standards-for-3...@googlegroups.com <community-standards-for-3...@googlegroups.com>
Subject: Re: [3D Standards] Fwd: [lib.dig.pres] Archival formats for 3D objects?

 

To view this discussion on the web visit https://groups.google.com/d/msgid/community-standards-for-3d-data-preservation-cs3dp/CA%2BQWpq6zLg0%3DHMDG4gkgW2y493btP6yHH4pBXPmwy1QGSHVu%2Bg%40mail.gmail.com.

[Polys, Nicholas]

I should also mention that X3Dv4 can compose and present glTF models 

 (in a similar way that webpages compose and present images, text)... 

So, for example, 

https://examples.x3dom.org/gltf2/

and

https://www.kshell.com/pages/webinar_20220531/drinking/goblets_inline_html.html

NB:

There are text AND binary encodings of both X3D and glTF

br,

_n

---

From: community-standards-for-3...@googlegroups.com <community-standards-for-3...@googlegroups.com> on behalf of Polys, Nicholas <npo...@vt.edu>
Sent: Tuesday, June 25, 2024 9:44 AM
To: community-standards-for-3...@googlegroups.com <community-standards-for-3...@googlegroups.com>; Kieron Niven <kieron...@york.ac.uk>

To view this discussion on the web visit https://groups.google.com/d/msgid/community-standards-for-3d-data-preservation-cs3dp/MN2PR05MB670401282A9724033FABB2CBB5D52%40MN2PR05MB6704.namprd05.prod.outlook.com.

[Murray, Kate]

Hi all,

 

Just chiming in from the rafters to say that I run the Sustainability of Digital Formats site at LC as well as co-organize the Library of Congress Recommended Formats Statement. We have entries for many of these 3D formats and are always happy to take comments and suggestions.

• 3D and Design Formats - Sustainability of Digital Formats
• Design and 3D - Recommended Formats Statement

(oh hey – I just realized that those two related things are not labeled in the same way. Great! I’ll see about fixing that…)

 

Best from Kate

 

Kate Murray (she/her)

Sustainability of Digital Formats

FADGI AudioVisual Working Group

Recommended Formats Statement

Digital Collections Management and Services

Library of Congress

km...@loc.gov

 

 

 

 

From: community-standards-for-3...@googlegroups.com <community-standards-for-3...@googlegroups.com> On Behalf Of Polys, Nicholas
Sent: Tuesday, June 25, 2024 10:01 AM
To: community-standards-for-3...@googlegroups.com; Kieron Niven <kieron...@york.ac.uk>
Subject: Re: [3D Standards] Fwd: [lib.dig.pres] Archival formats for 3D objects?

 

CAUTION: This email message has been received from an external source. Please use caution when opening attachments, or clicking on links.

 

Virus-free.www.avg.com

To view this discussion on the web visit https://groups.google.com/d/msgid/community-standards-for-3d-data-preservation-cs3dp/MN2PR05MB6704947075F6024441CE5B12B5D52%40MN2PR05MB6704.namprd05.prod.outlook.com.

[Thomas Flynn]

To summarize file formats in this thread so far that have been mentioned as suitable/possible options:

• OBJ
• PLY
• STL
• X3D
• X_ITE
  
• glTF
• 3MF
  

(Maybe I missed others?)

For some more context, a 2019 survey of cultural heritage organisations uploading 3D to Sketchfab surfaced that 62% of ~1000 organisations were uploading in OBJ format. Full results here, much discussion could be had around what these numbers mean, of course.

 

To return to Scott Prater's original conundrum (receiving files in OBJ or STL format), I think the answer may simply be 'both' (if you can), relying on the duplication of data to increase chances that files can be opened in the distant future. 

As many people have mentioned so far, different file formats support different things and it should be understood that it is fairly trivial to between most formats mentioned using open source tools like Blender or Meshlab.

A decision is also affected by the expected audience(s) and use case(s) for the data, working back from there might help guide thinking. 

Best regards to all,

Tom

Thomas Flynn

Digital Cultural Heritage Consulting

nebulousflynn.substack.com | LinkedIn | X | Sketchfab

ᐧ

ᐧ

To view this discussion on the web visit https://groups.google.com/d/msgid/community-standards-for-3d-data-preservation-cs3dp/PH8PR09MB10095D7F5C21E7281B6AA246ECBD52%40PH8PR09MB10095.namprd09.prod.outlook.com.

[Susan Borda]

Thanks everyone for the very informative responses!