LAPID: Using 3D to Recover Heritage Lost in a Fire

The Beginning

At the beginning of the 2000s, an informal partnership was established between researchers of the Paleovertebrate and Egyptology sectors of the Museu Nacional, the Divisão de Desenho Industrial of the Instituto Nacional de Tecnologia (DIVDI/INT/MCTIC), and the Clínica de Diagnóstico por Imagem (CDPI). Initially, the team focused on the acquisition of high-quality digital models of fossils of turtles (Podocnemis sp.), crocodiles (Mariliasuchus amarali, Stratiotosuchus maxhechti), and dinosaurs (Staurikosaurus pricei) for research purposes. Technologies like 3D scanning, Computed Tomography (CT scan), 3D digital modeling, and rapid prototyping were used.

After successful initial experiences, the team decided to expand to other types of subjects: Egyptian mummies. Three items of the Egyptian collection of the Museu Nacional were digitized using a CT scan: the never-opened coffin of Sha-Amum-em-su, a human child (which, after the CT scan, was revealed to be a cat), and the head of The Beauty from Thebes (an unidentified mummified head that has been part of the collection since the beginning of the XIX century; the results of the CT scan revealed it was a young woman, but her identity remains a mystery).

These experiments culminated in the creation, in 2002, of the Laboratório de Processamento de Imagem Digital (Lab for Processing of Digital Image – LAPID) as part of the Departamento de Geologia e Paleontologia (Geology and Paleontology Department) of the Museu Nacional from the Universidade Federal do Rio de Janeiro, Brazil.

In the years following its creation, the LAPID team started to diversify its areas of application. The Museu Nacional began to 3D digitize objects from its diverse collections, such as meteorites and vertebrate skeletons, including a humpback whale that was part of one of the museum’s exhibitions. New partnerships were established and, today, our main partners are the aforementioned DIVDI/INT and the Núcleo de Experimentação Tridimensional (Tridimensional Experimentation Lab) from the Department of Arts and Design, Pontifícia Universidade Católica of Rio de Janeiro (NEXT/PUC-Rio).

Purpose

Despite being part of a geology and paleontology department, the LAPID has always developed diverse projects in partnership with other departments of the Museu Nacional and other institutions. Because of that, the purpose of the lab goes beyond the acquisition of 3D models of fossils and rocks. Hence, the purpose of the labs is to use and develop 3D technologies for research and conservation of biological, paleontological, geological, and archeological heritage (and any other topic that comes to us). Parallel to these applications, the lab also eventually uses 3D technologies for artistic projects.

Staff

The LAPID team is composed of researchers from different areas of specialization: biology, geology, archeology, and design. The team includes Ph.D., master, and undergraduate students who develop all kinds of projects, such as 3D morphometrics, paleoneurology, paleopathology, biomechanics, heritage conservation, kinematic studies of animal locomotion, and dance.

Technologies and Projects

CT Scan

One of the first technologies to be employed since the origin of the lab was CT scan because it allows non-destructive studies of the internal parts of collection items. For the paleontologists of the team, CT scanning allows the reconstruction and description of the braincase, inner ear, and aerial spaces located inside the skull of fossilized specimens of diverse species of vertebrates, such as crocodiles, turtles, and sloths. CT scanning was also used to digitize a block of rock brought from the field even before it was opened and mechanically prepared. After the digitization, the sediment could be digitally filtered and removed through a process that we call “virtual preparation”. This digital process revealed an almost complete skull and mandible of a new species of crocodile that was later named Pepesuchus deiseae. The “virtual preparation” subsequently aided the mechanical preparation of the fossil.

Surface Scanners

We also experimented with different surface scanners. The first 3D scans were made with table scanners: a Roland Picza LPX250 laser scanner (from DIVDI/INT) and, subsequently, a NextEngine, our first 3D scanner acquisition. It worked well for several small objects that could be moved from the museum collections and exhibitions to the lab. However, it was limited because there were a lot of large and delicate specimens that could not be moved or accommodated by such scanners.

So, our experiences with portable scanners began in 2009 with the HandySCAN 3D (from DIVDI/INT) that was used to digitize the Sha-Amum-em-su coffin as well as the Bendegó meteorite (Brazil’s largest, weighing 5 tons), and archeological items, such as Marajoara pottery (the Marajoara was a pre-Columbian era society from the Marajó island at the Amazon River. They produced large and elaborately painted pottery with representations of plants and animals). Hand scanners proved to be a good type of equipment for our projects because of their versatility, but this particular scanner required that we place several markers over the surface of the objects we were scanning, which wasn’t possible in several situations. Then, in 2011 we acquired a white light Artec MHT scanner, which produced our best results overall; some highlight examples that were digitized with Artec MHT are the fossil of a complete skeleton of the crocodile Mariliasuchus amarali and the complete skeleton of a Humpback whale that was in the main exhibition of the Museu Nacional.

More recently, LAPID borrowed an RV Scanner from NEXT/PUC-Rio. This equipment had a very high resolution and could digitize small specimens very fast. It was employed on several Ph.D. theses that required a large number of high-resolution models.

The digitization of the Bendegó meteorite with the HandySCAN 3D was done in several parts that were digitally merged (below).

Kinect

Experimenting with new things has been part of LAPID since its origin. Several times we had the demand for producing 3D models of people for artistic projects. The Artec 3D MHT that was in use at this time was very good for acquiring skin details, but subtle movements of the human models made the entire process very unproductive. Also, this scanner and others that we tried were very poor at capturing hair and beard. Surprisingly, a piece of very cheap equipment produced the best results: Microsoft Kinect. We opted for that equipment every time we had to digitize people. In addition, Kinect also was used at interactive exhibits to demonstrate to the public how 3D scanning technology works.

Virtual 3D modeling

Since the beginning, we also use virtual 3D modeling in several projects. One of the first research projects produced at LAPID (2005-2014) was the study of vertebrate locomotion. The fossilized skeleton of two basal dinosaurs, Staurikosaurus pricei and Saturnalia tupiniquim were 3D digitized and then completely reconstructed using virtual 3D modeling. The musculature of their hindlimbs was also reconstructed and compared to that of recent birds, Struthio camelus and Rhea americana. A kinematic study of these large birds using high-speed cameras, combined with 3D digital models of their skeletons, also helped us to understand how the bone morphology relates to the movement of their hind limbs during terrestrial locomotion.

 

From Photos to 3D

In 2009, LAPID developed the DinosVirtuais pilot project. The objective was to produce 3D models of several items of the Paleovertebrate collection and showcase them on a virtual exhibition that represents the architecture of the Museu Nacional building. Visitors from all around the world could visit the museum and navigate through the rooms on an experience similar to a first-person game. The 3D models needed to be simple and light enough to be quickly loaded from the internet, which had a very limited speed at that time. Also, the generation method needed to be simple and cheap because the objective was to gradually expand the exhibition with new items and new collections. At the time of execution of the project, generating a 3D model from a 3D scanner sometimes required several days of processing, so, an easier and faster method was used: shape-from-silhouette. This was the first time the team employed a 3D methodology that is based on conventional 2D digital images. Results were adequate for the intended use but were not adequate for scientific research and heritage conservation, which require high-quality 3D models.

Promo video of the DinosVirtuais pilot project, developed in 2009:

In 2010, researchers from the Ilhas do Rio (Island of Rio) project, developed in partnership with the Mar Adentro institute, came to us with the challenge of generating high-quality 3D models of an entire archipelago: the Cagarras Islands Natural Monuments, located to the south of the Ipanema Beach, in Rio de Janeiro (see more). None of the techniques that were in use at LAPID at that time were adequate. So, the team experimented with photogrammetry for the project. The team took hundreds of photos from a boat and combined them with frames extracted from videos recorded from a helicopter to produce a model of each island. The photogrammetry-based 3D models of the islands were combined with a bathymetry model of the seafloor in order to reconstruct the entire area of the archipelago. The results were surprisingly good compared to those obtained from 3D scanners. So, the team decided to gradually implement photogrammetry into new projects.

Cagarras Islands Natural Monuments by LAPID – Lab. Processamento de Imagem Digital on Sketchfab

For several decades, researchers from the Paleovertebrate Sector of Museu Nacional have been studying a Cretaceous outcrop popularly known as “Tartaruguito” that is rich in extremely well-preserved fossils of turtles and crocodiles. The team also found fossilized eggs and other kinds of material on this site. Usually, the team produced traditional 2D drawings to map the position of each fossil on the site. But, after the excellent results obtained at the Cagarra Islands, beginning in 2015, we decided to use photogrammetry again to produce a series of 3D models of the outcrop year after year. During the excavation process at “Tartaruguito”, several 3D models where generated, aiming to generate a complete virtual representation of the spatial disposition of the fossils inside the rock. This project is still ongoing, and the final 3D model will allow understanding of the taphonomical process of the site and goes beyond the traditional 2D maps sketch.

The examples described above show the versatility of the application of photogrammetry for the researchers and students of LAPID. Also, the portability and low costs of the required equipment (a digital camera, a set of lights, a computer, and software), made this the ideal methodology for the Ph.D. students that needed to travel to other collections around the world. Actually, two Ph.D. students are using photogrammetry as the main methodology for their projects. One of them is generating 3D models of cervical vertebrae of Pleurodira, a group of turtles, to understand biomechanical aspects and the morphological evolution of these bones. The other student is generating 3D models of crania and mandibles of South American native ungulates to understand how this group diversified during its evolution.

Photogrammetry became vital to us after the fire that struck the Museu Nacional and, subsequently, the LAPID. All our equipment (digital cameras, 3D scanners, computers, etc.) was destroyed during the fire. Photogrammetry was the easiest, fastest and best option that we had to continue our work.

Prototyping Technologies

Because of our partnership with INT, since the origin of LAPID, we had the opportunity to try different types of prototyping technologies. Initial results at INT using a ZCorp zPrinter 310 were so good that, in 2011, we decided to buy one for our lab. Shortly after, we also acquired a colored model, the ZCorp zPrinter 250. For the first time, we could produce, for example, a digital endocast model of a fossil crocodile, such as Mariliasuchus amarali, and produce a physical copy of it. We were also able to reproduce in actual size the entire skeleton of a cat mummy from the Museu Nacional’s Egyptian exhibition. The mummy was tomographed and its prototyped skeleton was then incorporated into the exhibition. Also, for the first time in Brazil, we prototyped a digitally reconstructed entire skeleton of a dinosaur, Staurikosaurus pricei.

Our prototypes have been showcased in several national and international exhibitions, such as the 3D Print Show (London, 2012) and 3D Imprimindo o Futuro (Museu de Astronomia e Ciências Afins – MAST, Rio de Janeiro, 2017).

After the Fire

Following the fire that destroyed the Museu Nacional main building in September 2018 (follow this link to know more about this tragedy), all the knowledge about 3D technologies built by the LAPID team was put to use. These techniques are being used to help the researchers and staff of Museu Nacional to rescue and preserve the affected collections.

We have been acting on two fronts. The first is the digitization of internal areas of the palace and architectural elements, because the building is historical heritage from the Brazil’s imperial period. Digitization of internal areas is also serving to guide the systematic rescue of fragile scientific collections items, being a record of the artifacts’ arrangements before removing them from the site. The second front is the 3D digitization of the artifacts recovered by the rescue staff in order to record the current condition of the artifacts before they go through the restoration process. The database of rescued 3D models will enable the researchers to access the artifacts virtually, avoiding the necessity of handling them.

From Ashes to Ashes

After the fire, the LAPID team thought they could do something to honor the collection items and parts of the museum main building that were totally destroyed and carbonized. These materials could not be saved anymore. So, the team partnered with researchers from NEXT/PUC-Rio and a new unconventional research branch was established in LAPID: 3D printing of scanned collection items of the Museu Nacional using a blend of materials collected from the museum ashes.

Initially, the blend was composed of the original printer material mixed with ashes from the wooden coals obtained from the leftover burned remnants, but we are currently experimenting with different combinations of material. In 2019, the first results were shown to the museum’s staff and to the general public during the 201st anniversary of the Museu Nacional. On the occasion, the team presented the 3D print of the skull of Luzia (the oldest South American human remains) and the heart scarab amulet of Sha-Amum-em-su (3D digitized from the rescued original amulet). Both these prototypes were printed using debris and ashes obtained from the rooms from where they were recovered. This exhibition brought an emotional connection between the people and the Museum’s history, represented by these important pieces made from the ashes.

Museu Nacional rescue by LAPID – Lab. Processamento de Imagem Digital on Sketchfab

Books and the Future

In an attempt to disclose and popularize the application of 3D technologies in “STEAM” (“Science, Technology, Engineering, Arts, and Mathematics”), LAPID and partners decided to compile their initial research projects in a book entitled 3D Technologies: Palaeontology, archaeology, fetology (Revinter, 2009). Two new books have been produced since then.
The second book, entitled 3D Technologies: Unveiling the past, shaping the future (Lexikon, 2013), expanded on the first, compiling research from several new areas: Egyptology, Archaeology, Paleontology, Medicine, Biomimetics, Biophysics, Meteoritics, Biology, Geology, Fine Arts, and Design. The third and most recent book, entitled Seen Unseen: 3D Visualization (Rio Books, 2019), compiled research in areas as Medicine, Biology, Fine Arts, Archaeology, Paleontology, Engineering, and Mathematics.

Our three books show a little of what has been done in LAPID and at our partners’ labs.

Those books showcase what the LAPID team has been doing. Its importance to the Museu Nacional recovery and reconstruction can be easily observed from the models that we are sharing. If we could summarize the almost twenty years of existence of LAPID, we could define our lab as a multidisciplinary experimentation image lab that explores the boundaries of 3D technologies for research and conservation of cultural heritage.

At the moment we are slowly trying to reconstruct LAPID after the fire, searching for resources, donations (you can help us!) and acquiring new equipment. We may have lost all our equipment, but we haven’t lost either the ability to use and develop new technologies nor the ability to support heritage preservation at the National Museum. We will slowly recover from the ashes, but the road ahead is still long.

Visit LAPID on Sketchfab