Technological Revolutions and Art History, Part Three: Carla Schroer and Mark Mudge
- Okay. Since we are a couple of minutes behind and I wanna make sure we have time to get to a question and answers at the end of this. We are gonna come back and start with a presentation from Carla Schroer and Mark Mudge, co-founders of the nonprofit Cultural Heritage Imaging known as CHI.
CHI develops and implements imaging technologies for cultural, historic, artistic heritage and scientific research. They've also been great friends and collaborators with MoMA. So I'm excited to turn it over to them for their presentation, bringing scientific imaging to a broad base of cultural communities. - Hi. Thanks for all of you coming
to listen to this interesting process today. And I'd also like to thank Lon Addison for setting us up so beautifully. So let's get started.
I'm gonna begin and then Carla will take it away after we set the stage. At CHI, we are committed to the idea and what brings us into work every day, that the treasures of humanity are worth saving. And CHI's vision is one where simple to use, easy to learn documentary technologies are widely available to cultural heritage practitioners, cultural communities, including indigenous communities, and citizens scholars around the world.
And what we're truly committed to is preparing the ground for cultural communities to take control of the documentation of their heritage. But more importantly, to take control of their own cultural narratives so that others from other places in the world, and frequently ex colonial powers, aren't providing the narrative for these communities. We also want the information that's built by citizen scholars, cultural communities, and heritage professionals to be reliable and accessible to people around the world when the cultural communities give permission for sharing. We also wanna lay the groundwork such that conservation and preservation for both the original material, piece of material culture is enhanced, but also the documentary digitization of that material is also preservable and enhanceable going forward because Lon accurately pointed out that the digital dark age has consumed a great amount of early cultural heritage documentation and we'll never get it back. So, the important point is that if we set the foundation appropriately, digitizations or investments in digitization with scarce cultural heritage dollars are going to pay dividends in re-use of this information going forward. One of the most important concepts in this context is the democratization of technology.
And one of the most important components of this is allowing the work of cultural and cultural communities and citizens scholars to be accepted by the world community. And what we need there is to be able to first simply capture the scientific metadata that describes the context of gathering digital information, and then the processing of that digital information into digital representations. But we need to be able to separate the authority and empirical reliability at these digital representations from the institutional authority.
We wanna separate the authenticity of the digital representation from the institutional authority of the group usually a first world group frequently former colonial operators, from only authoritative sources being able to contribute to the cultural dialogue. And so what we want is for anyone who learns a good photogrammetric computational photography workflow to be able to contribute their work to the world's knowledge base and have others qualitatively evaluate that work. So they can understand whether it's appropriate to reuse or repurpose. So democratization requires a set of technologies where the equipment to capture the stuff in the real world is accessible.
Meaning it's relatively low cost and the learning curves involved in understanding how to do these processes are manageable and short-term. And that's because for us to do any good in trying to mitigate the disaster of cultural heritage loss, particularly with the increase of the destruction, of climate change, we need these procedures to be adopted widely throughout the world. And so, if we can get cultural communities and citizen caretakers all over the world involved in this documentary process, we're going to enrich humanity's knowledge base. And this activity will also subvert the ongoing process of digital colonialism by well-financed first world societies.
When CHI went to Nigeria sponsored by the US State Department, and I went with Eric Lansburgh formerly director of media at MoMA and one of Robert's early mentors, the Nigerians were absolutely determined that taking control of their own cultural narrative was central to the work of their monuments and museum operation. And the audience here is completely full of Nigerians from their national museums. And so, here's an example, we were at a national museum in the city of Ibadan Nigeria North of Lagos, and I was given a tour of this beautiful museum that has all sorts of stuff that I, as an outsider, would have selected to digitize. But during our three day seminar there, I said "go out out and find an object that is most important to you and we'll digitize it". So, the local museum community came back with an object that I would never have selected in a hundred years to digitize.
And it was this percussive instrument here. It's clay and it's got an opening at the top and an opening in the middle, which was used in song and dance in the local community which is incredibly important to Nigerians. Another project that we've been involved with is bringing the skills of computational photography to the Tlingit community in Southeast Alaska. And the last thing we did before COVID shut us down was go to the British museum on behalf of the Maori community in New Zealand and digitize a sail that had been collected in the early 1800s, which is the only example of this type of canoe sail that exists.
And the community of weavers in the Maori group wanted to recapture the weaving technology that enabled these sails to be resistant to weathering and highly efficient when you're piloting a canoe. So, we used photogrammetry using a nine position redundancy process to create a 3.6 billion point point cloud of the sail. And here's an example of an ortho mosaic of the 3D digital representation that we created. And the model was too big to efficiently be used on the web right now. So, we use digital elevation maps and ortho mosaics to make it usable. And now I'm gonna transfer the con? over to Carla.
- Thanks Mark. So really what we wanna state here is the big idea is that we know how to use off the shelf digital cameras and low cost software to create all kinds of digital representations that can be very high quality, very high precision if you do them correctly. And I think that if you do them correctly, is part of the goal of the digital lab notebook which I'm gonna to talk about in just a moment.
So our work at Cultural Heritage Imaging is all in the field of computational photography. And what we mean by that is that instead of taking an individual photograph, we're taking sequences of photos and we're using computer algorithms to extract information across that sequence of photos so that we get a new kind of digital representation that's not possible from a single image by itself. And there are many, many examples of this.
Our organization primarily works with Reflectance Transformation Imaging or RTI, and I'm gonna show you an example in a moment, photogrammetry for 3D and some spectral imaging. So, a quick example here of a 3D model, this is from the Smithsonian. And just to show you, this was part of a training class, it's part of their anthropology collection. So the blue rectangles are the photos. And basically we have, you know, and I just modeled off this piece, top and bottom, then we can zoom in and this is all high precision measurable data. I'll show you very quickly in RTI.
And the idea of an RTI here is that, So, for 3D we were taking a lot of photos around the subject, and then we have a full 3D model that we can manipulate and rotate. For RTI, we have a single camera point of view, but from that single camera point of view we have shape and color and we can look at really fine surface details and we can relight however we want to really bring out those details. So, we also can mathematically enhance the surface. So, in this case, by taking the color out and making the surface kind of shiny, we can see every little crack.
We can see the little bubbles that were probably part of the original manufacturer. We can see some of the corrosion that's happening on the surface. Just another quick RTI example. This is also from the Smithsonian, and then here's an image from the RTI. And then here's an image from that enhanced mode. And we can see every little chip, every little touch and retouch on the surface.
We have a ongoing project in Albania right now where we're we're working with the Albanian-American Development Foundation and also the ministry of culture of Albania and helping them set up a center, a national center for digitization. And we'll be doing photography, but also RTI and photogrammetry and teaching local members of the museum community over time how to apply all these technologies. So this was part of the pilot project we shot this beautiful icon, and, here's just a quick version on Sketchfab. So, we actually shot this in both photogrammetry and RTI, and you can see the surface and lots of cool detail. And then here, this is from the RTI, just images directly out of the RTI.
You could see the level of surface information that we have. So, we can really understand both how it was made, its current state, from a conservation perspective, and often we can also see evidence of prior conservation activity when we look at these really fine surface details. So, back to this idea of democratization of technology, that's really what drives our organization. And one of the central ideas there is that local communities need to be in control of what gets digitized, what gets shared. There are reasons they may wanna keep things just within their community, and there are things that they wanna share.
And that the more we involve local communities and empower local communities, the more that all of us are enriched and they are empowered to be part of the narrative and the story that's being told. So how do we do this? So, the first big idea Mark gave you. Here's the second big idea, which is that we know how to collect and manage and create a digital lab notebook that records everything that we're doing, who, what, when, where, how and why, and of the full context so that when people are doing this kind of digitization, they can have appropriate records that other people can inspect and understand what they're looking at. So, this is open source software that we're working on right now. There's an early beta available from our website.
In this short presentation I can't give a demo, but there are videos on our website that talk more about it and how you can use it. We have a grant right now from the National Endowment for the Humanities, and we'll be releasing an updated beta in the next month or two and then by June, we'll have that one data release out. The big idea is it's open source, we try to make it easy to use. I think a critical thing is that you can reuse data. So there's a database on the backend. You enter information about your equipment, your people, your stakeholders, your projects, whatever, and then you can kinda group that together and make templates and reuse that data really easily.
It's also super flexible. So you can put a huge amount of information in, you can put a small amount of information in, and that's up to you and your institution. We're producing linked data that's mapped to the side of Conceptual Reference Model which is a semantic ontology. And what's really cool is you don't need to understand anything that I just said. It just does this for you, it works under the hood, and we think that's critical for people to adopt the current kinds of technologies and semantic knowledge that's happening out there.
The other thing is that the tool not in the early beta but that the next version that's coming out, we have our an archival tool that allows the user to automatically wrap up their original images, this metadata, work products that go with it, processing information, all of that, and put that in either a METS or a Bagit format, making it much easier for people to submit their information to repositories. So the idea here is we're collecting and managing information about the digital representation through its whole life cycle, and then preparing that data to make it easy to submit and manage and put in repositories. So, the kind of information that we're collecting is all kinds of things from, you know, equipment, who was involved? We can associate documents with the information, who paid for it? We have rights information, what's the subject? I mean, our focus is on the information about the digital representations, but we need to link that to what actually is this a representation of. So, and the idea with linked data is then as there's other linked data out there, you can go find other references to that particular site or subject on the web. Right now, we support Reflectance Transformation Imaging, photogrammetry, multi-spectral imaging and documentary image sets.
So it's such a very photo image set oriented system. A couple of quick acknowledgements about the folks that have been involved. And we're a primary partner here is the center for Cultural Informatics in Herklion Crete. We've had funding from a variety of sources over the years as we've been exploring these ideas and now from the National Endowment for the Humanities to really build it out.
And I just wanna end here by noting that there is a lot of information on our website. So the RTI software is free and open source. You can download it from our website, and there are instructional videos and user guides and all kinds of cool stuff. And similarly with photogrammetry, we right now are using commercial software but there are instructional videos to help you with that.
The digital lab notebook software is open source and there are instructional videos and guides and things that go along with that. So we really encourage you to take a look if you're interested in pursuing any of these technologies. And I'm gonna end it there. Thank you.