Deciding on the features your model will prioritize
Determining your access to the model’s subject and data about the subject
Choosing a workflow and deciding on tools
Developing the model
Exporting the model into outside programs for animation, annotation, or other types of sharing
While you may go back and forth at a few stages in this process, especially as you develop a model and then try to use it for further research or presentation, you do not want to return to the earlier stages after you have moved into developing a dimensional model in a specific tool. As few scholars in this field have experience building dimensional models before working on a project like this, you will likely need to allow significant time to learn the tools used for modeling. While you can certainly work with experts, these skills are not as common as those like web coding, and the costs for expertise can be high. Such collaborations also require that you understand the strengths and weaknesses of the tools available for modeling different aspects of historical or cultural subjects, as the workflows and skills required are different. It is very difficult to switch to a different workflow when you’ve already invested in a different process. For this reason, expect to spend a lot of time outlining your project’s priorities and required features, and thinking realistically about your access to the subject and to source data.
What is your project’s purpose? (Be as specific as you can be.)
The first concern in building a dimensional project is delineating a clear purpose for your project. It is very common for dimensional projects to be started with the vague idea that the result will somehow be engaging to a wider audience, or the desire to show that you are using cutting-edge technology. However, this lack of specificity won’t help you plan and will likely lead to an end project that doesn’t meet your needs, that is just a digital ghost of your subject. Lack of a clear goal can also cause you to spend a lot of time (and sometimes money) trying to model an aspect of your subject that isn’t necessary. You should not expect to produce a simulacrum of your subject, and each tool and technique for modeling prioritizes different aspects of buildings or objects.
Why do you need to create a model? What will your project gain from the modeling process?
Start by assessing what a digital model of your subject can do that can’t be done with the original subject itself, and the reasons why the original has those limitations. This might be related to research questions, like wanting to be able to manipulate a fragile object to see how its parts work, to test reconstructions for damaged material, or to “place” a digital model into a reconstructed environment. Perhaps you want students to construct a model as a classroom assignment to encourage them to analyze the object or building in a new way. The goal for the digital model might also be related to presenting the subject to an audience in a way that is impossible with the original. This is common with very small or very large subjects that are hard to grasp in person, or simply with artifacts that can’t be placed on display or that you want to make available to people who can’t be there in person. Or you might wish to show an audience several phases of a building over time, or to show a building as it looked in an earlier time.
After this, you should be able to state your goal in a single sentence. If you find yourself with several goals or multiple things that the model should do, you may need to consider breaking the project into different models, or at least different phases that might occur in separate tools.
What do you want your model to represent or show about your subject?
Consider which aspects of your subject material are the most important to this goal. Is it scale, materials, relative dimensions, surface effects, relationship to the setting, the structure of moving parts, or how its assembled? It is something that only relates to the object or building itself, or will you need to incorporate aspects of a larger setting or environment? While you will likely have more than one answer to this, try to narrow it down and to note the things that are less important. For example, you may decide that you are most interested in volumetric or geometric aspects of a building, and that you won’t need to model elements like wall color or materials of construction. Or, if you are trying to see whether fragments fit into a missing part of a sculpture, scale and surface wear may be particularly important. If you are thinking about visual effects, you may want to focus on lighting and medium. Your goal here is to narrow down your list of features to the minimum number that can be used to explore your questions or to convey your subject to its intended audience.
Then, consider your resources, especially your access to the subject and source data about it.
Choose a modeling process:
There are two main processes for producing dimensional models. One starts with the subject and combines large numbers of photos of the subject from different angles to build a three-dimensional photogrammetric model. For this, you need access to the subject, not just to see it from a few angles, but from every angle, including the tops of buildings and the backs and bottoms of objects. You also need the object to be well-lit from all angles. It is very difficult to use this photogrammetric process if you are working from an object in a glass case or if your access to a building is all at ground level (though you can often create a model of part of a building this way, like a facade or single interior room). If you are working on a building, you may need to consider tools like drones or scaffolding to take photos from high angles. In some cases, you might be able to build a photogrammetric model from historic photos, but you would need a large number of photos from different angles, so this is rarely a promising path. If you have the necessary tools and access to the subject, photogrammetry is an excellent process for creating very detailed models that have a lot of surface variation, variable curves, or ornament, like complex building facades, intricate sculpture, organic objects, or fine metalwork. There are other processes for starting with an extant building and creating a digital model, like laser scanning (LIDAR), but these require specialized resources and training beyond the scope of this handbook. Very complex architectural structures and ruins are often scanned using these processes.
The second process for producing dimensional models uses measurements and known data about the subject to create a model. This process is necessary if you do not have sufficient access to the subject you want to model to take photographs from all angles, or if it no longer exists. In some cases, this can also require fewer resources: generally no physical equipment is required to create basic models, though you may need specialized applications and you may need additional resources to present the model in some forms.
In addition to the pragmatic reasons you might need to use this process, you may choose this process if you are especially interested in scale, volumetric parts of a building or object, or the way different parts of an object or building fit together or move. Since you are modeling each part of the subject yourself, the more complex your subject, the longer it will take you to create the model. It can be very fast to model a simple box-like building or spherical object this way, and if you have the right data, you can work remotely using only a computer and some software to process the model. Although you won’t want to model complex geometry this way if you can avoid it, some tools allow you to add photographs or other 2d image files to the surfaces in your model to create realistic surfaces, while other tools can simulate different materials. This is also a good route if you need to animate parts of your model or show change over time. The critical thing to remember with this process is that you need data about the subject that is complete enough to substitute for having the subject itself as a model. For every project, you need to know proportions (and probably absolute size) of all the different parts, often including aspects of depth that can be difficult to find in published material. In some projects, you may need other information about the subject’s interior, medium, or other features that may not be clear in photographs.
How will the model be used? Who is your audience?
Finally, you need to think about how the final model will be used. Sometimes, there is a research question and the creation of the model itself is all you need. In most cases, however, you will want to put the model online, create animated movie clips, or at least export still images from different angles for publication. Common output or presentation goals for the creation of dimensional models include:
Exporting snapshots of your model from different angles.
Creating an animation or video that guides an audience through your model as part of a story.
Printing a 3d model of your subject.
Embedding an interactive virtual model in a website so that other users can manipulate your model themselves.
Creating a virtual or augmented reality experience for a wider audience.
Exporting the model for use in another tool, like a GIS platform, for further manipulation.
If you need your audience to be able to manipulate parts of the model or to experience it in a self-guided or virtual reality type project, that will both limit the tools you can use to build the model and require your audience to have access to special software or equipment like VR headsets. This limits your potential audience and adds to the likely cost and time to develop your model, so consider carefully which aspects of presentation are critical for your audience’s needs.
When you have a clear idea of your goals and the features of your subject you wish to model, you can begin to evaluate tools and develop a timeline for production.
Continue Reading: Tools & Resources for Dimensional Projects