Possibilities and constraints for virtual visits: Experimental approaches to VR at the Smithsonian American Art Museum

Sara Snyder, Smithsonian American Art Museum and Renwick Gallery, USA


Starting in 2016, the Smithsonian American Art Museum undertook a series of experiments in producing Virtual Reality (VR) applications, in order to gain hands-on experience with the technologies involved, and to increase the museum staff’s firsthand knowledge of the medium. The objective of the projects was to evaluate different types of VR—from relatively lightweight mobile apps to robust, room-scale experiences—in order to understand the strengths and weaknesses of each as a vehicle for delivering educational and aesthetic experiences to a remote audience, in keeping with an art museum’s mission. This paper presents and analyzes two primary case studies from among these experiments. The first case study is SAAM’s in-house production of “Renwick Gallery WONDER 360,” a VR app for mobile devices, created as a compliment to the 2015 WONDER exhibition at its Renwick Gallery. The second case study, “Beyond the Walls,” is a more ambitious, room-scale VR experience for the HTC Vive system, and was developed with support from our commercial partner, Intel. Each of these case studies centers around an application meant to offer a virtual visit to a section of the art museum galleries, and together they give a glimpse into the possibilities for virtual museum visitation in the future. The paper analyzes the costs, benefits, and lessons learned in each project, and suggests an approach and set of criteria for evaluating future developments in the VR space that could be utilized by museums both large and small.

Keywords: VR, virtual reality

Virtual Reality in context

Industry and media hype over consumer Virtual Reality (VR) has reached a fever pitch over the last four years—ever since Facebook acquired VR startup Oculus on March 25, 2014 for a reported $2 billion. Facebook’s commitment to VR catalyzed new investment within the technology industry, and kicked off a VR arms race that sent rival companies like Google, Sony, Microsoft, and HTC into overdrive in the production of their own VR systems. Only a few months later, at the 2014 Google I/O developers conference, Google Cardboard headsets—notable for their simple, low cost construction—were given away to all attendees. In 2015, shortly after the Oculus acquisition, both Time and Wired magazines published cover stories on VR, raising mainstream awareness of the recent advances in the technology.

VR has had enthusiastic proponents for many decades. The concept of stereographic head-mounted displays datse to the 1960s, but entered common vernacular in the 1990s, thanks largely to science fiction and popular culture (Earnshaw, 1993).  For example, in 1992, VR pioneer Jaron Lanier and his company, VPL Research, created the real VR headsets shown in the movie  The Lawnmower Man. And while the intervening decades saw no commercially successful entertainment VR, it has been a standard technology for prototyping industrial designs and training surgeons for the last twenty years (Lanier, 2017). But finally, after 2014, it seemed possible that VR was indeed on the brink of changing the consumer entertainment and education landscapes as well.

To be sure, skepticism still abounds, but there is no doubt that the head-mounted displays (HMDs) that enable a truly immersive VR experience are getting faster, smaller, and cheaper at a rapid pace. A mass market VR tipping point, beyond the limits of the hardcore video gamer market, could be only a few short years away; one of the main ingredients that will drive consumer interest in VR is the creation of more compelling types of use cases and content outside of gaming. In 2015, prices began to drop on consumer-grade 360-degree photography and video cameras, putting the ability to create immersive images and videos within reach for many consumers and museums. Facebook and YouTube added full support for 360-degree video feeds, and the NY Times and Washington Post began regularly publishing 360-degree features for mobile phone VR viewing.

It was within this context that the Smithsonian American Art Museum’s (SAAM) Media and Technology office began in earnest to experiment with VR in 2016. Our staff had a great deal of interest, but no real hands-on experience producing VR experiences; nor did we have much budget for equipment or outside contracts with VR production firms. However, we did have two very important advantages. First, we had our leadership’s support for innovation and experimentation; SAAM’s director and deputy director were not only open to the museum running VR experiments, they provided pockets of seed funding for software and equipment. On an even larger scale, the Director herself cultivated relationships with industry partners in tech and entertainment via the museum’s board members. This drive towards innovation at the very top of the organization proved crucial. Second, our work was building upon a longer-term commitment to cultivating in-house media and technology talent. Our museum’s media team is fairly small, but we had spent years encouraging training and conference attendance for our staff, including our media producer, Carlos Parada. Instead of sticking to the typical museum conference circuit, Parada had been asked to the South by Southwest (SXSW) Interactive conference. His exposure to the ideas and trends of the larger tech industry at the height of the VR hype cycle in 2015 and 2016 meant that he was versed enough in the landscape of the industry to advise us, and help us pursue connections with partners outside of the museum world. When it came to VR, he was able to recognize what was truly an innovative opportunity, and what was just hype. While the museum couldn’t have known he would play that role for us, we had a general faith that investing in staff conference attendance would pay off—and it has.

These two enabling factors are simple enough to explain, but not easy to attain. Many museum directors might say that they support innovation, for example, but the truth is that where an organization spends its energy and money ends up being its highest priority. For an art museum to seek both board members (and their friends) from the world of technology may take the museum out of its comfort zone, but the board and the leadership end up the stronger for it. In addition, it might seem simpler or less expensive, at least in the short term, to contract out every new technology project and need; technology staff commands higher salaries and can be difficult to recruit to the museum world at a time when they are in high demand in the private sector. Training and travel, similarly, can be among the first things to get cut in tight budget years. Yet each investment in a technical staff person builds upon the knowledge and expertise that museums can bring to bear on evaluating and experimenting with an emerging technologies.

Virtual Reality in museology

The published discourse in museum literature about the substitution of digital simulacra for material objects dates back to the early days of the Web. However, there has been relatively little focus within the museum literature on VR experiences that fully immerse the user and place them inside the virtual world (as opposed to just looking at or manipulating a 3D simulation on a screen). For the purposes of my research, I am limiting my definition of VR to the types of fully-immersive experiences enabled by such devices as head-mounted displays and haptic wearables such as controllers or gloves, and do not consider all 3D virtual environments built for computers as true “virtual reality.”

In 2002, Klaus Müller wrote optimistically of the potential of VR for museums: “Just like film and theater, VR applications broaden the way we perceive the world. However, in a VR environment the visitor becomes part of the virtual world and can change it through his actions. VR allows a person to use her mind, eyes, and hands to enter a place she may have otherwise only visited in her imagination” (Müller, 2002). Despite this enthusiasm, the realities of the marketplace meant that we didn’t see any substantial case studies in real-world museum applications of fully immersive VR until over a decade later.

In more recent years, a survey of VR discussions in the museology literature reveals an increase in the case studies after 2014, as would be expected in our post-Oculus/Facebook world. As Juno Rae and Lizzie Edwards have documented, beginning in 2015 numerous museums began capturing their spaces in 360-degree panorama, and offering them as mobile VR experiences through Google platforms such as the Google Cultural Institute and YouTube (Rae, 2016).  In the summer of 2015, the British Museum hosted a Virtual Reality Weekend event, the success of which led them to conclude that VR offers both educational benefits to the visiting public and a chance to pursue interesting partnership opportunities (Rae, 2016). A 2016 project at the Canadian Museum for Human Rights included both remote and on-site availability of their VR experience, Weaving a Better Future. Launched in coordination with the opening of the exhibition, it helped audiences to connect to the artisan women featured in the exhibition by virtually transporting them to the artisans’ home environments in Guatemala, via Samsung Gear VR head-mounted displays (Gillam, 2017).  At Museums and the Web 2017, Michael Fischer argued for the democratizing power of VR, as he led participants through a thought exercise in launching a museum that solely housed native VR artworks, a step in preparing the museum community for that future possibility (Fischer, 2017).  One engaging recent example of a non-native VR artwork brought to life—indeed, it is over 450 years old—is a component of the digital-only exhibition focused on Pieter Bruegel the Elder, sponsored by Google Arts and Culture and eight different European museums. Bruegel: Unseen Masterpieces includes a virtual reality experience for Google Cardboard in which the fantastical creatures depicted in Bruegel’s Fall of the Rebel Angels come to life, with interpretive narration provided by the Royal Museums of Fine Arts of Belgium (Caessa, 2016).  Several exciting museum VR announcements came to light in 2017, including the British Museum’s partnership with Oculus to create a virtual tour of some of their most popular galleries, and a multi-million dollar investment by HTC in the VIVE Arts Program, which aims to “work with museums and content developers to bring their works to life in a whole new way” (British Museum blog, 2017; Gepp, 2017).

Distinguishing degrees of VR: Mobile vs. desktop

Before presenting the two specific SAAM case studies, a distinction needs to be made between the technologies under discussion. The term VR is used to describe what presently falls into several types of mainstream commercial experiences, which are distinguished along a spectrum of cost and complexity by the degrees and types of positional tracking, freedom of mobility, interactivity, and frame rates they deliver (among other factors). For the sake of simplicity, as a shorthand I will refer to the two common endpoints on the spectrum as “mobile VR” and “desktop VR.”  While there is an emerging category of middle-tier devices (such as Oculus Go) poised to hit the marketplace in the near future, and an overall trend away from tethering to powerful desktop machines, for the (brief) present moment, I believe this division accurately describes the landscape.

Mobile VR

EXAMPLES: Google Cardboard, Samsung GearVR
PROS: Powered by your smartphone (or a smartphone-esque device), this is an easy, inexpensive way to enjoy 360 photos and video.  Untethered by wires or cables, it uses the positional tracking and app capabilities built into the mobile devices that people already own. The cost of the HMDs is very low, and the cost of producing 360-degree assets is modest.
CONS: Only three degrees of freedom (3 DoF) of mobility mean your feet are essentially planted, and you are limited in exploring the world from different angles. They usually have little interactivity, due to limits of processor power and the positional sensors in mobile devices. There are no additional hand controllers. Downloadable VR apps tend to be very large, and eat up phone memory. These devices are more likely to cause motion sickness in some users; they are also by far the type of VR most popular with museums at the moment, because of their lower costs and low barriers to entry.

Desktop VR

EXAMPLES: HTC Vive, Oculus Rift, PlayStation VR
PROS: These can be used for all of the experiences built for mobile VR, plus games and experiences which include “rooms” or spaces that have been fully modeled or rendered in a 3D engine. This means that a user has a full six degrees of freedom (6 DoF) and is able to move and explore in all directions, either by physically moving their body, or by using the system’s hand controllers to transport themselves. They are less likely to cause motion sickness than the 3 DoF devices.
CONS: Today, these HMDs still must be tethered to a computer or some type of high-powered gaming system, and they rely on externally-mounted position tracking sensors (this will only be true for a short while longer, as fully-contained, untethered head-mounted devices are due on the marketplace soon). The hardware and software are not mature, and therefor can be buggy or get overheated. Currently, each HMD costs $400 or more, plus the cost of a machine (computer or gaming system) powerful enough to run its related software. In addition, creating a fully-navigable 3D world is expensive and requires a high degree of creative skill and resources. Museum assets for such a system would have to be captured with high resolution scans and photogrammetry.

SAAM case studies

For the purposes of this paper, I will not focus on VR-enhanced exhibitions where headsets are distributed within the physical museum building, as interesting as they are; rather, both of my case studies focus on the possibilities of the virtual museum experience, one that is created with the intention of it primarily being experienced in a remote physical location, away from the museum’s real-world footprint. I will consider the potential as well as the limits of the current state-of-the-art consumer VR to deliver on those two key value propositions that our museums offer: education, and inspiration. These are our strengths; how might they translate to someone who is physically far from our doorsteps?

SAAM case study #1: “Renwick Gallery WONDER 360”

Honestly, when it came to the creation of the Renwick Gallery’s popular 2015 WONDER exhibition, a VR app was never part of the plan.

interior of Renwick Gallery during WONDER exhibition, showing rainbow colored strands of thread that form Plexus A1, by Gabriel Dawe
Figure 1: Gabriel Dawe, Plexus A1, 2015, as seen in WONDER at the Renwick Gallery. Photo by Ron Blunt

Starting in 2012, the curator and the rest of the museum’s staff had been completely consumed with the logistics of reopening a renovated building and presenting the most ambitious series of large-scale contemporary installations the museum had ever hosted. The media and technology team were launching a brand new Drupal website built especially for the occasion, while also laboring over more than a dozen in-house video production projects, including artist interviews, as well as producing online components ranging from slideshows to a new e-commerce ticketing solution. We were swamped. By the time the museum and the show opened, in November 2015, the staff was ready to take a deep breath and relax for a minute; little did we realize that opening would be just the beginning of another fast-paced year. WONDER proved to be the Renwick’s most popular exhibition to date, attracting more than 730,000 visitors from November 2015 through July 2016 and creating a sensation on social media and with the press.

As it happens, the year that WONDER opened to the public was also the year that VR hit the mainstream. Throughout 2015 and 2016, the technology to support 360-degree panoramic capture got better and cheaper, month by month. Colleagues, visitors, and commercial vendors began to reach out and ask if they could come over and photograph WONDER in 360 for VR.  Some simply wanted to capture the experience for their own interest; others wanted the museum to hire them for the photography, which seemed both expensive and logistically difficult to manage, given the unrelenting crowds. The Renwick was packed with people all day during nearly all daylight hours, as well as after hours during evening special events. Was a clean, well-lit 360-degree photographic capture of all nine installations even possible, during such hectic times?

As mentioned in the introduction, in March of 2016, SAAM’s media producer Carlos Parada attended the SXSW Interactive conference in Austin, Texas. While there, he connected with a startup company, InstaVR, and he saw the kind of opportunity we were looking for: an innovative, cloud-based software system that would enable us to publish a VR tour quickly and relatively easily. Furthermore, Parada felt confident that he could do the photography work himself, provided we augmented the museum’s existing Canon EOS 5D Mark III DSLR camera with some new equipment and software. The museum’s director agreed to let us try, and we purchased the following items:

  • GigaPan EPIC Pro robotic camera mount, essential for the automated capture of panoramic photographs (cost: $995)
  • Canon 35mm prime lens (cost: $999)
  • Autopano Pro 4 software, to facilitate stitching the panoramas together (cost: $120), with additional editing done in Adobe Photoshop
  • Extra camera batteries, essential when shooting time is tight (cost: $200)

So, with only a few months left before the show’s closing, in April of 2016, Parada and the team worked up a plan that would result in the first ever VR app published by the Smithsonian. Our content goal was to present a virtual re-creation of the physical installations, allowing users to freely enter and explore each gallery in the building, in any order they might choose. Our naive metric of success was simple: with the clock ticking down toward de-installation, could we even pull this off? Our goal was learning the limits of the technology, our museum’s bureaucracy (the app had to go through a centralized IT security and privacy review process), and ourselves.

While there were many relatively inexpensive 360-degree cameras coming onto the market in 2016, we quickly deduced that the quality and resolution of the photographs taken with these machines was simply not sufficient to see the detail of the artworks. Using the new accessories for our Canon, Parada took hundreds of photographs of the entirety of the WONDER exhibition during the months of May and June, from a variety of vantage points. He stitched together a composite of 72 to 80 pictures for each final shot, using a combination of Kolor’s Autopano Pro and Photoshop. Our generous colleagues at InstaVR upgraded SAAM’s account to their Pro level, free of charge, and it was on http://www.instavr.co/ that he edited and integrated the assets using InstaVR’s cloud-based admin console, turning the 360-degree photographs into a virtual tour of the Renwick Gallery. He embedded the open-captioned artist interview videos that we had previously created in each virtual gallery space, inserting the artists’ own voices into the experience. Within the app, spaces within the Renwick Gallery are made navigable via a hands-free gaze cursor—a small dot in the center of the screen which selects the destination once you have focused the selector on top of the icon for several seconds. Finally, a subtle audio track was added, recorded in the actual gallery, consisting of echoing footfalls and a low murmur of voices in the background. While the audio is not an essential component to enjoying the VR, it can potentially increase the sense of being immersed in the space, conveying the sense of the scale and acoustic qualities of the materials used in the architecture.

Screenshot showing white arches along central atrium of Renwick Gallery, with silvery light sculpture in center
Figure 2: screenshot within “Renwick Gallery WONDER 360” VR app

The final product was a pair of native mobile apps for iOS (iPhone and iPad) and Android which were released on the iTunes App store and Google Play under the name “Renwick Gallery WONDER 360.” I wish we’d been able to launch the apps earlier, so that they could have been promoted during the run of the show, but that would have required a lot more prescience—and bandwidth—than we had available at the time. We also assumed that the mainstream popularity of iTunes and Google Play would give us the best exposure to the marketplace, but we will likely publish to VR-specific stores in the future, since the GearVR store seems to have been the largest source of remote downloads for the VR project at the Canadian Museum of Human Rights (Gillam, 2017).

At around 3,500 users to date, our Google Analytics data shows the VR users to be a mere sliver compared to the website and in-person traffic the show generated; partly that is due to platform choices and lack of marketing, but it certainly is also due to timing. However, now that the physical WONDER is long over, perhaps the VR experience has even more value as an archive.  As one iTunes app store reviewer wrote, “I missed this exhibit at the Renwick and was pleased to be able to see it using VR.” A reporter for the Washingtonian, writing about the app, characterized it as “a second chance at experiencing ‘Wonder’” (Whiting, 2016).  I also found it gratifying when a discussion of WONDER 360 found its way onto the fall 2017 syllabus of Johns Hopkins University’s Museums & Mobile course, taught by Eric Longo. The continued relevance of the project as an independent, virtual experience is rewarding, as it the fact that now, having done it once, it will be much easier to plan for 3D capture the next time.

More notable to me than the quantity of our app users is their remarkable geographic diversity. In the last six months, only one third of the apps users have been in the United States; the second and third most visiting countries are Japan and India. While we need more than the limited app store data currently available to make an informed analysis of what is driving the international downloads, they are a tantalizing reminder of the global nature and potential of the app marketplace to contribute to interaction long after a physical exhibition has closed.

SAAM case study #2: “Beyond the Walls”

In the fall of 2016, just after the WONDER exhibition closed and our first VR app launched, a SAAM board member connected with our then-Director, Betsy Broun, to brainstorm some ideas around virtual reality. What ensued was a meeting between industry experts in that board member’s circle—entertainment producers, technologists, 3D photographers—in which they were given the Director’s in-depth tour of SAAM. After 27 years of leading the museum, Broun had become a consummate tour guide, and by the end of the experience, the guests were wowed—both by the art, and by her stories. They commented that her tour was one of the most engrossing museum experiences they had ever had. How could the level of access and expertise that they’d just enjoyed in person be made available to anyone? Working together, could we offer a similarly compelling insight into to the Smithsonian’s collections through VR?

Over lunch, we toyed around with ideas that took advantage of the VR medium, and the fact that real world problems like physics or registrarial protections need not hinder us. Could we let people rummage through our storage spaces, or flip over every painting in the collection to look at the back?  Could we shrink users to the size of ants, then let them climb up our sculptures? Could we animate a painted scene from the Civil War, bringing it to life? Could we build the first art museum on Mars? While I expected us to head towards building a creative experience together that defied the constraints of our physical space, to my surprise, our guests were most excited by the idea of capturing and rendering the museum’s architecture in 3D, and then reinstalling and enhancing interpretation of a selected set of artworks for virtual visitors. In other words, they wanted to create a realistic, alternate version of the real-world Smithsonian American Art Museum. And why not? Housed in Washington, D.C.’s historic old Patent Office Building, it is an architectural gem and a designated U.S. National Historic Landmark. Add that to the fact that the physical Smithsonian museums are a site of pilgrimage for so many American schoolchildren, and it means that carrying that the real-world weight of stone and brick and history into the virtual space might not be a drag on creativity, but a means of harnessing the location’s unique power. It also means that a VR rendering can, potentially, serve as a meaningful substitute experience for a person who longs to travel, but cannot. Those virtual visitors can tell themselves, “this is what the real museum looks like,” and maybe feel less like they are missing out.

After several more discussions in the weeks that followed, we found ourselves batting an agreement back and forth between lawyers for Intel, the main industry partner, and for the Smithsonian. Intel had a wish to invest in an educational VR project that could showcase the power of their processors, and wanted to create a demo experience for their booth at CES 2017.  SAAM brought to the table the Smithsonian brand name, plus our artworks, educational content, and a staff enthusiastic about experimenting with VR, but little funds for such experiments. We found our way to an agreement that felt like a win-win, and within weeks, 3D photography expert Greg Downing and producer Peter Martin had returned to D.C. They began to capture our permanent collection gallery spaces and artworks using LIDAR scanning, focusing on an area of the museum’s second floor that offered spaces of a modest footprint and (important!) a collection of older artworks that were free of copyright and restrictions. The media and technology staff gamely signed up for shifts to come in a 5:30 a.m., on both weekdays and weekends, to escort our partners through the spaces so that they could work before the museum opened. Downing and Martin also visited Rock Creek Cemetery four miles away, to capture a compelling outdoor scene that complimented one of the sculptures. In less than a week, the capture process was complete. Next came the tricks of settling on a final checklist of artworks to feature, and gathering additional assets that would add value and context to the collection objects within the VR. Downing worked to combine the LIDAR scans with photogrammetry to build a high level of detail. The Smithsonian’s 3D Digitization team provided data for a sculpture they had previously scanned, The Greek Slave, and shared it with the technical production team. Under Martin’s direction, 8i provided volumetric capture and Framestore produced the Unity build, which made the rooms fully navigable. Four of the museum’s artworks were selected for special treatment and a little bit of VR “magic,” detailed below, using sections of the interpretive script for each work developed in collaboration with SAAM curators Eleanor Harvey and Karen Lemmey. Those artworks are as follows:

Frederic Edwin Church, Aurora Borealis, 1865 (https://americanart.si.edu/artwork/aurora-borealis-4806)

In this painting, Frederic Edwin Church has taken the aurora borealis—ethereal, dynamic, and alien—and captured it in static oil paint, making you believe that you are standing underneath that phenomenon, witnessing the colors reflected off the ice. In VR, you can stand closer to the painting than guards might ever permit in real life, practically putting your nose against it to examine its texture and observe its custom frame. If you’ve never experienced the real aurora borealis, it is hard to understand how the natural phenomena compares to Church’s scene. But VR users standing in front of the painting can trigger a teleportation hotspot which sends them to a remote mountain in Iceland, where they are suddenly in a dark landscape, looking around at jaw-dropping, 360-degree 6K video footage of an actual aurora blazing in the sky, provided by designer and photographer Olafur Haraldsson. The ability to compare and contrast the two scenes offers rich opportunities for learning and observation.

Augustus Saint-Gaudens, Adams Memorial, modeled 1886-1891, cast 1969 (https://americanart.si.edu/artwork/adams-memorial-21528)

In 1885, Marian Hooper “Clover” Adams, an amateur photographer and the wife of the writer Henry Adams, committed suicide by drinking poisonous chemicals used to develop film. Her grieving husband commissioned prominent sculptor Augustus Saint-Gaudens to create a memorial that would express the Buddhist idea of nirvana, a state of being beyond joy and sorrow. Saint-Gaudens modeled a powerful shrouded figure, and then worked closely with architect Stanford White, who designed a secluded, contemplative setting for Clover’s gravesite in Rock Creek Cemetery in Washington, D.C.. Decades later, the Smithsonian American Art Museum acquired a bronze cast made after the original in the cemetery. When a user is standing in front of SAAM’s bronze cast, they can choose to teleport to Clover’s actual gravesite, where they come face-to-face with the same sculpture, but this time in the context of the private outdoor memorial for which it was originally intended. Soft sunshine filters through a bank of trees, which move softly in the background, and the bench surrounding the sculpture allows for a moment of quiet contemplation. Flashing back and forth between the museum’s version and the outdoor version, you can notice the differences, sometimes subtle, that distinguish the two casts, and the effects of weather on the outdoor installation.

Screenshot showing back of nude female sculpture, The Greek Slave
Figure 3: screenshot within “Beyond the Walls” VR experience, The Greek Slave in foreground, Aurora Borealis to the left, and Adams Memorial against the far wall.

Hiram Powers, Model of the Greek Slave, 1843 (https://americanart.si.edu/artwork/model-of-the-greek-slave-20060)

The original marble sculpture of The Greek Slave propelled its artist, Hiram Powers, to international stardom. The Greek Slave was almost immediately associated with the anti-slavery movement in the United States, as abolitionists used images of it to promote their cause. The 3D model that appears in the VR app was rendered from a scan of the original plaster model that dates to 1843; in fact, this VR edition is a not work that exists in the real world at all. The presence of this sculpture in VR provides an opportunity to draw parallels between contemporary 3D scanning technology and nineteenth-century mechanical reproduction techniques, and to talk about the slippery (and often unhelpful) concept of “the original,” when it comes to sculpture.

Alex Prager, Face in the Crowd, 2013 (https://americanart.si.edu/artwork/face-in-the-crowd-110667)

The only contemporary artwork to appear in the “Beyond the Walls” VR experience is a selection from a video installation by Los Angeles-based artist Alex Prager. When you experience it in the physical museum, Face in the Crowd is installed in a black box gallery, where video plays asynchronously on three of the walls. The experience in VR looks no different, with one notable exception: the artist herself is standing in the room with you. You can walk up to Prager (or around her—she was volumetrically scanned in 8i’s Los Angeles studio, and has been fully rendered in three dimensions) as she tells you about the inspiration for her artwork as you experience it “together.”  Prager’s work deals with the anxiety of being swept up by the masses while trying to create and maintain a sense of self—conditions long present in the physical world—and how this anxiety can be amplified in the virtual spaces we inhabit today, making it a very fitting piece for VR indeed.

Screenshot showing navigational grid along simulation of the museum's wooden floor
Figure 4: screenshot showing navigation using a hand controller within “Beyond the Walls” VR experience


Table holding computer, monitor and HTC Vive headset and controller
Figure 5: demonstrating the “Beyond the Walls” VR experience in the Smithsonian Castle along with Intel, at the Smithsonian Digitization Fair, October 2017. Shown are an HTC Vive HMD and controller, and HP backpack-style VR machine

While “Beyond the Walls” is not yet available to the public as of this writing, the Smithsonian and Intel are working out the agreement that governs the next phase of the project, and everyone on both sides is excited to see it released as an experience that will likely be available on the HTC Vive and Oculus gaming systems. Having spent many hours over the past year watching Smithsonian staff and board members testing out the experience in room-scale VR, I’m excited to continue down this path. I’ve never seen such smiles as when visitors discover they can fly into the aurora borealis and back again, to see it in a 19th century masterwork, with renewed context. (To learn more about the industry partners involved in the production, see Framestore’s video at https://youtu.be/GHoS1ysjMG0).

Drafting an evaluative checklist for future VR projects

Like all tools, VR’s value depends entirely on what objective the museum is trying to achieve. Now that we have these learnings to draw from, for future projects at SAAM we are asking ourselves a series of questions:

  • Does our goal present a problem that VR is uniquely qualified to solve? We don’t believe in VR for VR’s sake, but sometimes, it is just the right fit. For example, in October of 2017, the Renwick Gallery opened a new exhibition, Murder is Her Hobby: the Nutshell Studies of Unexplained Death. The show featured a series of exquisitely detailed dioramas, constructed by Frances Glessner Lee in the 1930s and 40s as training tools for the emerging field of forensic science. Each diorama depicts a mystery, and their interiors work well for up-close scrutiny. Yet the dioramas are small, extremely fragile, and had to be exhibited behind glass. Building on our knowledge of 360-degree capture honed during WONDER, we once again used InstaVR to create a VR experience, this time supported on the Web and an in-gallery media kiosk, rather than requiring app download. With VR, visitors can enter the dioramas as if they were doll-sized themselves.
  • Can we do the VR well? Do we have the time and budget for it?  If we have to do it in a low resolution, or in a rush, it might not be worthwhile. Are there educators/curators who are willing to devote energy to the project’s content, not to mention editors, producers, and designers?
  • Does the VR add value for the visitors? Can we build in user experience testing or surveys that will enhance our knowledge of how VR is perceived by visitors, including metrics such as satisfaction, enjoyment, and learning outcomes? Are there any barriers, such as usability or accessibility?
  • Can we learn from it…and maybe even enjoy the process? Even if the project doesn’t necessarily require VR to be successful, does it give the museum a chance to stretch our wings, building and growing our own staff’s knowledge and experience?  Would it be an interesting experiment? Does it captivate our imaginations? Would it be (gasp) FUN?
  • Do the benefits of doing another VR project outweigh the costs? Benefits include value to the visitors, but also the chance to connect and network with valued technical partners, such as we found with Intel, VALIS studios, and xRez. Are there ways that the costs can be reduced, such as doing some of the 3D or photographic capture in-house? One conclusion that we have reached with VR, as we have with a number of other digital experiments, is that in an effort to keep costs down, the museum won’t commit to bending over backwards trying to maintain a project forever. All new digital projects are assumed to have a three-year timeframe, at which point we evaluate the costs and return on investment of maintaining public access. If we can keep it going for another three years, securely and reliably, we commit to doing so.  If we cannot, we take steps to archive the project and sunset it.


These are still early days for mainstream, consumer VR, and no one yet knows which companies and platforms are overhyped, or which will turn out to be major commercial contenders. But we do know for certain that virtual worlds will become increasingly easier to access as hardware prices drop and displays become lighter and untethered from computers. We can also be sure that the social aspects of VR will increase through the presence of avatars, and that meeting friends or classmates in virtual worlds is likely to become an important feature. For museums, websites and social media didn’t replace our brick-and-mortar buildings: they enhanced our ability to connect with people around the globe. VR is likely to follow the same pattern. A world where a classroom full of student avatars tours the Smithsonian, following a guide and asking questions, cannot be far away. Like the Web before it, VR will not make SAAM’s physical buildings or original objects obsolete, but will make a richer experience available to an ever-larger population, regardless of their ability to travel to Washington, D.C.

The evidence for the impact of such immersive 3D experiences is promising. A recent meta-analysis found evidence that games, simulations, and virtual worlds “were effective in improving learning outcome gains” (Zahira, 2014).  Other research indicates shown that 3D museum tours, specifically, “appear to be more effective for engaging learners through a realistic-looking environment,” and that “the greater the similarity of an online museum experience is to its physical counterpart, the more positive will be the observer’s perception of the collection” (Katz, 2015). This seems a promising indication that we should continue creating and testing experiences that leverage the familiarity of museums as physical spaces.

But will industry solutions for creating and publishing commercial VR meet the needs of museums? Perhaps, but only if we continue to proactively partner as content experts, and to let the technology vendors know what museums—and our virtual visitors—want and need. At MCN 2017, in a discussion panel about VR in museums, SFMOMA’s Keir Weinsmith expressed his view that early partners from cultural heritage have a rare chance in the current moment to shape the direction of where this emerging industry is headed with VR (Evjen, 2017). I am hopeful that he is right, and look to my colleagues in the field to pursue strategic relationships with industry and present them with an ever-increasing number of museum use cases, like those described here, as well as many more.


The work described in this paper would not have been possible without the good work of the following people: Carlos Parada and the SAAM Media and Technology team, Betsy Broun, Rachel Allen, Stephanie Stebich, Nathaniel Phillips, Ed Fries, Lauryn Guttenplan, the InstaVR team, Rajeev Puran and Intel, Greg Downing and xRez Studio, Peter Martin and VALIS Studios, Framestore, 8i, Eleanor Harvey, Karen Lemmey, and Alex Prager.


British Museum blog. (2017). “New Virtual Reality tour of the Museum with Oculus.” Published April 18, 2017. Consulted January 13, 2018. Available https://blog.britishmuseum.org/new-virtual-reality-tour-with-oculus/

Caessa, P. (2016). “When Bruegel met Google: immerse yourself in a masterpiece.” The Keyword. Published March 15, 2016. Consulted December 16, 2017. Available https://www.blog.google/topics/google-europe/when-bruegel-met-google-immerse_81/

Earnshaw, R. A., M. A. Gigante, and H. Jones. (1993). Virtual Reality Systems. San Diego: Academic Press Inc.

Evjen, M., A. Andrea Montiel De Shuman, E Gangsei, R. W. Owen, B. Dawson, S. Fernandez-Duque, & K. Winesmith. (2017). “Best Practices for Creating Meaningful AR/VR Experiences.” MCN 2017. Published November 9, 2017. Consulted January 11, 2018. Available https://youtu.be/auj3a-V_O6Y

Fischer, M. (2017). “A proposal for a virtual reality museum for virtual reality art.” MW17: MW 2017. Published February 2, 2017. Consulted January 14, 2018. Available https://mw17.mwconf.org/paper/a-proposal-for-a-virtual-reality-museum-for-virtual-reality-art/

Gepp, M. (2017). “HTC VIVE Announces VIVE Arts Program.” Vive blog. Published November 8, 2017. Consulted January 13, 2018. Available https://blog.vive.com/us/2017/11/08/htc-vive-announces-vive-arts-program/

Gillam, S. (2017). “Spotlight VR/AR: Innovation in transformative storytelling.” MW17: MW 2017. Published February 28, 2017. Consulted January 14, 2018. Available https://mw17.mwconf.org/paper/spotlight-vrar-innovation-in-transformative-storytelling/

Hardawar, D. (2017). “The Smithsonian Art Museum Dove into VR with Intel’s Help.” Engadget. Published August 2, 2017. Consulted January 14, 2018. Available https://www.engadget.com/2017/08/02/smithsonian-art-museum-intel-vr/

Katz, J.E. & D.J. Halpern. (2015). “Can Virtual Museums Motivate Students? Toward a Constructivist Learning Approach.” Journal of Science Education and Technology, December 2015, 24 (6), 776–788. Available https://doi.org/10.1007/s10956-015-9563-7

Keshtcher, Y. (2017). “The Complete Guide For Web VR and Immersive Experiences.” UX Design Collective. Published August 25, 2017. Consulted January 29, 2018. Available https://uxdesign.cc/the-complete-guide-for-web-vr-and-immersive-experiences-220ce6fbb110

Lanier, J. (2017). Dawn of the New Everything: Encounters with Reality and Virtual Reality.  New York: Henry Holt and Company.

Müller, K. (2002). “Museums and Virtuality.” Curator: The Museum Journal, January 2002, 45, (1), 21-33.  Available http://dx.doi.org/10.1111/j.2151-6952.2002.tb00047.x

Rae, J. & L. Edwards. (2016). “Virtual reality at the British Museum: What is the value of virtual reality environments for learning by children and young people, schools, and families?” MW2016: Museums and the Web 2016. Published January 28, 2016. Consulted January 10, 2018. Available https://mw2016.museumsandtheweb.com/paper/virtual-reality-at-the-british-museum-what-is-the-value-of-virtual-reality-environments-for-learning-by-children-and-young-people-schools-and-families/

Stuart, P. (2017). “#VRWC17: How to put VR into Museums.” Preloaded.com. Published April 18, 2017. Consulted January 29, 2018. Available https://preloaded.com/vrwc-put-vr-museums

Whiting, A. (2016) “The Smithsonian Just Released a VR App of ‘Wonder,’ and It’s Beautiful.” Washingtonian. Published October 5, 2016. Consulted January 5, 2018. Available https://www.washingtonian.com/2016/10/05/the-smithsonian-just-released-a-vr-app-of-wonder-and-its-beautiful/

Merchant, Z., E.T. Goetz, L. Cifuentes, W. Keeney-Kennicutt, T. J. Davis. (2014). “Effectiveness of virtual reality-based instruction on students’ learning outcomes in K-12 and higher education: A meta-analysis.” Computers & Education, January 2014, v. 70, 29-40, ISSN 0360-1315. https://doi.org/10.1016/j.compedu.2013.07.033.

Cite as:
Snyder, Sara. "Possibilities and constraints for virtual visits: Experimental approaches to VR at the Smithsonian American Art Museum." MW18: MW 2018. Published January 15, 2018. Consulted .