Tactile Education and the Value of Apple’s Object Capture for Museum Educators
The importance of tactile experiences within museums and heritage sites has been recognized for some time. Hands-on, interactive displays seem to be on the rise, with the benefits of this type of experience being thoroughly studied and published (Wilson et al. 2018). Diverse museum audiences have been shown to reap the benefits of tactile engagement (Cooper 2019). Unfortunately, finding the time and money to create 3D models can be a daunting challenge faced by museum educators. The recent introduction of the Object Capture API by Apple, however, may be just what museum educators need to overcome those challenges.
Despite the benefits of tactile engagement, the decision to create such experiences within museums is not always an easy one. All museums provide two essential functions: (1) the conservation of their collections and (2) public education based on those collections. But this is often a balancing act. While handling museum objects is an effective teaching method, it is often harmful to the objects themselves since handling any object will inevitably cause cumulative wear and degradation. Many museums solve this dilemma through one of two methods–the creation of a separate “education collection” or the creation of printed 3D models of objects within their collection. Education collections consist of genuine objects that have been determined to have more educational value than research value for any number of reasons and, as such, can be handled by the public (MacFarlan 2010). These collections provide the unique benefit of allowing visitors to handle authentic collection objects; however, they have their own unique disadvantages that can be solved by using 3D replicas instead.
3D Models vs. Education Collections
Education collections are ultimately consumable collections, meaning they will eventually break and/or deteriorate beyond usefulness. 3D-printed models, on the other hand, are easily replaced through simple reprinting. Second, not all objects are suitable for education collections due to various reasons, including fragility, rarity, temporary existence (such as the short lifespan of a flower), or potential harm to the handler. Such objects can be replicated and printed as 3D models for safe handling. Lastly, objects in education collections cannot be altered. By contrast, museum educators can change the size and color of a 3D model. They can print the model using different materials; change its weight; or alter its surface with paint or labels. All of these variations allow museum educators to customize the object to fit their specific teaching needs.
However, creating 3D models presents barriers of its own that many museum professionals find difficult to overcome. These barriers can be summed up as the general lack of two key resources within a museum —time and money. Traditionally, learning the software used to create 3D models via photogrammetry has been a time-consuming process. Popular software is often highly complex, requiring practice and experimentation to master the basics. Additionally, the software itself often takes hours (or sometimes days) to render a single model. Secondly, the software commonly used to render photogrammetric models has been expensive. Museum budgets cannot always justify spending thousands of dollars on software alone. Luckily, a new application programming interface (API) developed by Apple called Object Capture practically eliminates these common obstacles.
The Benefits of Apple’s Object Capture API
Apple’s Object Capture “uses photogrammetry to turn a series of pictures taken on your iPhone or iPad into USDZ files that can be viewed in AR Quick Look…or used in professional 3D content workflows.” Object Capture greatly reduces the amount of time required to create 3D models. First, it only requires a few clicks to create a model. People new to photogrammetry no longer need intricate knowledge of point clouds to create a high-quality model, drastically reducing training in photogrammetric software. It also creates automatic masks, reducing hours of masking individual photographs by hand. Object Capture also reduces rendering times to minutes or hours instead of hours or days required by most existing photogrammetry software.
In addition to demanding less time, Object Capture also greatly reduces the cost of making photogrammetric models. Software that utilizes Object Capture, like PhotoCatch (used to create the models discussed below) and Easy Photogrammetry, are free to download, saving a museum educator thousands of dollars. This is great news for museums on a tight budget (aka most museums). With all of this in mind, Object Capture allows museum educators to produce quality models for hands-on education with relatively little time and money.
Using Object Capture to Create a 3D Model
While originally intended for the creation of models for augmented reality experiences, Object Capture can also be used to produce OBJ files (.obj) that can be used to print 3D models to be used as hands-on objects for museum education. The below model of a coyote skull was rendered using PhotoCatch in approximately 15 minutes. It consisted of 96 JPEG photographs (.jpg) taken using a light tent and Nikon DSLR camera. The skull was rotated 180 degrees (from teeth down to teeth up) halfway through the photoshoot to enable the photographer to capture images of all surfaces. (The full 3D model is embedded at the end of this post.)
The Limitations of Object Capture
With this in mind, Object Capture does have its unique drawbacks. It cannot eliminate the need for expensive hardware. As an Apple-designed API, Object Capture, unsurprisingly, only runs on Apple hardware. Additionally, being a relatively new creation, Object Capture only runs on the most recent Apple models — Intel Macs with 16GB RAM and an AMD CPU of 4GB VRAM or Macs that use the M1 chip (Esposito 2021). Newer models translate to higher price tags. And, on the subject of expensive hardware, we must include the purchase of a 3D printer or the cost of paying a third party to print the 3D models.
As mentioned previously, the coyote skull model pictured above was created using photographs taken using a DSLR camera. Although designed to work with smartphone cameras, a DSLR camera may still be necessary to produce models of certain objects when using PhotoCatch. To test the capabilities of Object Capture, a second model of the coyote skull was attempted using photographs produced using a Samsung Galaxy smartphone camera. Caution was used to ensure the same number of photographs were taken from approximately the same angles. This time around, PhotoCatch produced a highly inaccurate, and somewhat nightmarish, model.
Conclusion
Even with these limitations, Object Capture provides benefits to museum educators that far outweigh its downfalls. While hardware remains expensive, this does not present a total barrier for most museum educators. In mid-size and large museums, at least one of the collections or research divisions will already own a DSLR camera. And somewhere in the building, someone’s office is most likely equipped with an Apple computer. By working with other divisions, museum educators may be able to avoid making expensive purchases. And, in the case of 3D printers, the cost of small models is becoming increasingly affordable, selling for anywhere from $300 to $2000. To evaluate the full potential of Object Capture within museum education, additional research should be conducted. In the meantime, Object Capture’s ease of use, speedy rendering times, and minimal price tag make it an increasingly valuable tool for museum educators to create new and engaging tactile experiences for their visitors.
