Photogrammetry as a Digital Tool for Joining Heritage Documentation in Architectural Education and Professional Practice

Abstract

Given the accelerated technological advance in all fields, vast knowledge of digital tools becomes mandatory for future interior designers and architects. Thus, experimenting with as many such methods and technologies must become a priority for the teaching process. Noticing a high demand for digitalization during the COVID-19 pandemic, the Polytechnic University of Timisoara has decided to encourage the implementation of digital teaching throughout disciplines, preparing students for their future careers. Photogrammetry as a study method, among others, has the potential to outrank traditional documentation techniques currently applied in architectural education and practice. The presented research was divided into two main phases: an educational, experimental workshop and a project restoration application. After learning, testing, and refining the close-range photogrammetry workflow, the authors and students took the best practices and applied them to an ongoing facade restoration project in Timisoara, scanning original and restored plaster ornaments and finally 3D printing one of the models. The article aims to show that, unlike traditional teaching methods, using photogrammetry as a documentation process creates a coherent link between theoretical education and restoration practice. Consequently, the exercise brings students closer to the applicable side of their careers through a contemporary digital technique.

1. Introduction

Politehnica University of Timisoara (UPT) is going through an accelerated digitization process on all levels. The pandemic period was probably one of the strongest impulses to speed up this process of permanent change and development of the traditional school. The educational platform Virtual Campus is only one example of the tools that urgently adapted to the need of the staff and students when all classes were moved online.

Being actively involved in the context of complex transformations on all levels, UPT constantly adapts and is prepared to face and contribute to these challenges with an acute need for innovation, research, development, efficiency, technology, and most of all digitalization. Throughout more than 100 years of existence, UPT in collaboration with prestigious local and national universities and faculties, has contributed to a modern development, while maintaining a strong contact with the professional fields that offer constant feedback and support. Determined to further include digitalization in the teaching process, the Faculty of Architecture staff got involved in several external collaborations to better understand the potential of using photogrammetry in linking education and the professional field. Among the first contacts with the digital world of photogrammetry was right after a collaboration with one of the Erasmus partners schools during Design Week in Granada.

Following the participation in that active class in which the traditional method of photogrammetry was used in combination with a workshop for the recreation of some objects from various recomposed pieces (re-use), the authors came up with the idea of developing a practical work in our faculty as well, in which this technique aimed to help students from the Architecture and Design specialization in their future projects.

Over the course of the last year, most of the seminars within the Faculty of Architecture and Urban Planning have adapted to this digitization trend, either totally or partially changing the classic exercises, mainly using advanced 3D construction techniques, virtual reality visualization of projects, 3D printing, and 3D Scanning (photogrammetry) with the scope of creating objects on different scales by uniting these techniques. Some results of these new projects were disseminated through articles [1,2] and exhibitions with open access. Digital technology is recently being used in architectural restoration practice to create a greater connection between computer-generated data and actual site work, leading to a new trend in restoration, since it not only inherits culture but also improves accessibility, efficiency, and quality. Additive manufacturing (AM) mainly known as 3D printing, is a novel technology by virtue of which construction and building materials are being transformed over the last 30 years, towards a new reality that will soon replace the traditional way of construction. The most important part of the article addresses the procedure by which the authors managed to replace traditional construction methods and building materials through additive manufacturing and focusing on the repair and renovation of existing historical buildings.

1.1. Photogrammetry in Heritage Conservation and Creative Industries

1.1.1. Photogrammetry Definition and Principles

Since the invention of architectural photogrammetry by Dr. Albrecht Meydenbauer and the French officer and scientist Aimé Laussedat during the 1890s [3], photogrammetry was used in a wide range of domains, starting from archeology, topography, military applications, architectural conservation, and restoration, and more recently the gaming and film industries.

“Photogrammetry is the science of obtaining reliable information about the properties of surfaces and objects without physical contact with the objects, and of measuring and interpreting this information […]” [4]. Aerial photogrammetry, used for land surveys, was the first type of professional photogrammetry, which requires special cameras or drones. The basic principles of photogrammetry rely on stereographic restitution resulting in 2D elements, 3D models, coordinates, or topology [4,5]. Photogrammetry uses a variety of hardware and software for image alignment and mesh generation, including specialized laser scanners, and robotic devices, but also simpler less expensive methods that only require the use of digital cameras or smartphones, which are easily accessible to students. With several additional guidelines and principles, the hypothesis of the research was that inexpensive close-range photogrammetry has the potential of becoming an easily accessible analysis method, as well as a tool for professional use.

1.1.2. Close-Range Photogrammetry Scope

Technological development led to increasing digitalization in photography and photogrammetry and to the development of a considerable number of image processing methods. Digital image processing has spread across fields, including civil engineering [6], architecture, design, heritage conservation and documentation, industrial engineering, media, and computer-generated imagery. Having such an important impact across fields, integrating photogrammetry in architecture and design in higher education becomes mandatory. Consequently, during the first stage of this research, the authors sought to find optimal ways to introduce this workflow in education and observe the benefits and challenges it might have on teaching and further, on architectural heritage documentation.

The documentation of a cultural heritage implies acquiring, processing, presenting, and recording the necessary data for the determination of the position and the actual existing form, shape, and size of a monument or historical element in three-dimensional space at a given moment in time [6].

3D scanning is among the most recent methods of site surveying, specifically for historical sites and heritage where irregularities are present and, crucially, to be identified. The main purpose is the accurate and automatic detection of objects and data [5,6] that would otherwise be overseen or simplified by traditional surveying methods. Among the important uses of close-range photogrammetry is heritage damage detection and analysis [3] which might become a real tool for monitoring historical sites and documenting them for future restoration. When used for this purpose, close-range photogrammetry can be considered a 3D scanning method since it allows real data recovery and its transposition into virtual environments after several digital processes.

1.2. Photogrammetry as a Bridge between Education and Practice

1.2.1. Photogrammetry as an Educational Tool

Computer-assisted methods are increasingly developed in education and modern technologies are used enthusiastically by students [7] once they learn how to put the technology into practice. Universities are integrating 3D scanning, modeling, and visualization knowledge to identify, design, and evaluate an effective educational experiment [8,9]. Since methods are adaptable, students can be involved in all stages of the photogrammetry process [8] with minimal prior knowledge of handling digital tools.

There is also a social implication when studying heritage, as it encourages social implication in cultural events, respect towards history, and continuous learning while strengthening a sense of belonging [9]. This method has the potential to be part of Perspective or Descriptive Geometry seminars in our faculty, as all seminar applications are currently practiced in a traditional style (by analog technical drawing) [10] and this could be another opportunity to make steps towards digitalization.

1.2.2. Importance of Digital Techniques in Architectural Learning

Heritage observation, along with architectural theory and technical drawing, forms the basis of architectural education. Traditionally, observing valuable heritage details in school implies observational drawing, essays, or recreating replicas of emblematic architectural pieces through technical drawing or at most, 3D modeling. Theoretical disciplines give students knowledge for better identification, selection, and analysis of architectural heritage. The same fundamental process is required when choosing subjects for photogrammetry. Additionally, studying different textures, and surfaces and identifying valuable art and architectural elements in their proximity are already important steps in their career development.

Digital tools for architectural drawing and modeling involve an entirely different mindset compared to freehand drawing. Students are first taught how to run the software, create their own CAD templates and specific workflows, and then use the entire system to their own advantage in better design and collaborations. When referring to the photogrammetry workflow, there are clear indicators of the process sequence, but there are also variables regarding lightning, subject characteristics, or hardware, so the authors took advantage of these aspects to allow students to experiment, fail, observe, and improve. The exercise started with small-scale, simple objects and progressively advanced to more complex situations that required previously acquired experience [9]. Trial-and-error and “learning-by-doing” were basic teaching methods applied throughout the entire workshop: first intuitively, by observing and constantly changing or adapting and later deliberately, through a written report by each team of students. Enthusiasm for such a process increases due to anticipation, as results may vary depending on the quality and quantity of pictures taken. The excitement of each achievement, which is a clearly modeled and textured mesh, makes students determined to keep moving forward, finding ways to improve their photography method, and searching for new, optimal subjects.

1.2.3. From Theory to Practice Using Photogrammetry

Lack of coherence between theoretical principles learned in university and demands of architectural practice has been a major issue for alumni adaptation to the working environment in Romania. While teaching methods and technological facilities in Romanian higher education are largely the same for many years and the work dynamic has developed much faster, universities must envision updates quickly. In this regard, the paper explores an attempt to bring students closer to professional challenges, by raising questions and problem solving [8]. Such teaching methods are already successfully applied in universities worldwide, especially in the fields of heritage documentation and conservation, archaeology, construction engineering, and topography.

1.3. Thematic Justification

As the authors are practical work coordinators in the Faculty of Architecture for first- and second-year students, they could notice that although students show interest in office practice, they are rarely truly involved in real professional endeavors, due to a lack of digital experience. Oftentimes, when passing from analog to digital representation methods, young students show signs of discouragement due to the sudden rupture from the tangible world [7]. Evidence of the previous statement comes from our own didactic experience, directly experiencing the transition of the students from the exclusive use of analog tools of expression (such as pencil, brush, watercolor) in freehand drawing where they had absolute control, to digital techniques employing the computer or various devices repeatedly, that generates stress until the moment they manage to master it.

Although several hypotheses were brought to light at the beginning of this research, the one that shows the connection between academic and professional fields was followed throughout the entire process.

Photogrammetry is a complex method, as previously described; however, it can also be accessible to students who inevitably use smartphones and laptops but have still minimal knowledge of CAD and modeling principles. The exercise aimed to raise students’ interest in digital tools through experimentation while fostering a connection to real working environments. Thus, the authors argue that inexpensive, close-range photogrammetry has the potential of becoming an accessible method for students, as well as a tool they can rapidly apply in practice.

One important scope of the study was to bridge the gap between theory and practice and encompass all the involved actions since most academic activities remain mostly theoretical with hardly any contact with the professional world. The entire research follows this idea and only by further observing the development of this group of students over time, some clearer conclusions can be drawn on whether this intention was fruitful to their career, compared to other groups of students.

2. Methodology

The presented study was part of summer practice activities of the 2021–2022 academic year, in which Interior Design students from the Faculty of Architecture and Urban Planning of Timisoara engaged in as part of their curriculum. First-year students are often encouraged to choose representation and workshop-like activities compared to second-year students who already take part in design and architecture office endeavors. Thus, the practical work called “Representation techniques using methods of Photogrammetry” was well integrated into this curriculum trend, as well as the above-mentioned institutional digitalization intentions. Additionally, the authors aimed to create a link between the educational—experimental area and the professional restoration field as an important step towards application in a real restoration process. Consequently, the methods perfected during the workshop allowed students to be directly involved in the ongoing process of architectural detail documentation and reconstruction was taken one step further to the production of ornament molding, complementing the digitally generated mesh with 3D printing technology.

For most first-year students this was among the first interactions with digital techniques, as they practice freehand and technical drawing during their first year of studies. This allowed them to be introduced to the world of digital software through an unexpected, hands-on experience since photogrammetry implies transferring real-life elements to three-dimensional environments. The exercise was conducted over five days and involved six first-year students, grouped into three teams of two. The basic requirement and team formation criteria were the ownership of at least one laptop per team and the instructors provided information to download and install the Trial Versions of the software “Reality Capture” and a free full version of “Instant Mesh”. Furthermore, through a clear workflow, students learn and understand the architectural characteristics of historical buildings while learning 3D modeling and visualization techniques.

As part of the work methodology, the authors opted to divide the workshop into several clear steps, delivering a theoretical base (step 1), practical part (step 2), and software support (step 3), but simultaneously allowing students to experiment and discover their own techniques for photogrammetry. Following the outcomes of the workshop, the authors and students have applied the learned skills to an ongoing building restoration (step 4).

2.1. Theoretical Base and Guidance

This first step included explanations of the photogrammetry technique, professional uses, general workflow, photography principles [5], lightning principles, choice of items, and applications in the design and architecture field. The authors emphasized that a successful model generation relies on the number of photographs and strategy, as each point should be intersected by at least two rays of satisfactory intersection angle [4].

The instructors provided a theoretical introduction to the principles of the standard photogrammetry workflow that were described during the first meeting (detailed in

Table 1. Guidelines for the photogrammetry scanning technique (recommendations).

Subject Choice	Lightning Choice	The Actual Photography Method
Objects without shiny surfaces	Filtered light (choose a cloudy day)	For freestyle photography, turn around the object
Uneven/natural/textured surfaces are optimal	Create a uniform, filtered lightning when indoors	Take as many photos as possible in a previously established order (e.g., clockwise)
Objects must not have dark areas or holes	Glossy/mirrored or glass surfaces are to be avoided	Shooting levels must be successive from bottom to top or top to bottom in a “spiral” movement
No sharp edges/right corners		Rotate the object or rotate around the object as many times as possible
The color/texture of the object should be vibrant/varied		Take the photos at several heights to include the details below and above
		Photos must be focused, not blurry

), also giving general guidelines for subject choice, lighting, and photography tips. Students were presented with an overview of professional uses of photogrammetry in film, gaming, and other creative industries as well as heritage conservation and survey use, which were of bigger interest to our field. This session ended with a Q&A part when free discussions were encouraged regarding the transition from analog representation techniques to digital ones, CAD, and m software requirements in university and future careers.

2.2. Practical Part of the Workshop

Making the photogrammetry technique accessible to the academic environment meant choosing a set of easily attainable tools that did not involve using costly professional 3D scanners. This concept involved using handy devices such as personal smartphones and laptops, multiple lamps for artificial light sources, and various types of hand-crafted lightboxes. The instructors encouraged students to experiment with mobile applications and digital cameras instead of smartphones, but these devices were not the focus of the workshop.

The students started by building three different Light Boxes (Figure 1a–c) of various sizes, materials, and techniques, all proposed by the students based on previously viewed online tutorials:

• Light Box 1: cardboard structure with white paper sides and white paper cover (H66 × 53 × 53 cm)
• Light Box 2: cardboard structure with white paper sides (H43 × 25 × 53 cm)
• Light Box 3: wooden structure with white paper sides and baking paper (H72 × 70 × 70 cm)

Students continued with the photography process of small items of choice. The instructors intentionally allowed students to choose objects that would potentially not be suited for photogrammetry so that they could notice features that might not work.

The next step was the photography of small outdoor items: natural elements or small objects. Natural uniform lightning was a great advantage for this phase, as further described in the Results Section.

The final steps were photographs of architectural details: facade surfaces, doors, and sculptures. As the most important part, students passed one week of testing different surfaces to prepare for on-site work.

2.3. Software Support—Photograph Alignment

After obtaining an optimal number of photographs, depending on subject size and complexity, the next step was their insertion into the “Reality Capture” software and performing the “Align pictures—Draft” command. If the resulting point cloud is satisfactory, one can move to the “Reconstruction—Preview and Normal Detail” steps, obtaining the first images of the mesh and future 3D model. Students encountered certain problems at the alignment stage depending on the quality of the photos, the nature of the object, and lightning, and at the detailing stage, the problem of the performance of the graphics card intervened. It is mandatory to note that normal detail generation is possible only with “Nvidia graphic cards with at least 1000 MB memory and at least 2.0 CUDA compute capability”. Applying texture and mesh export in “.obj” format, are the last steps in the Reality Capture software. Before exporting, students were able to verify the object’s accuracy and narrow the view box to exclude unwanted environmental elements. These obstacles only raised productive questions for the students, and they managed to find solutions together with the instructors or even alone.

When trying to use the resulting models in 3D projects, the students noticed the large size of the files, hence the need to use an intermediate software, “Instant Meshes”, to optimize the mesh. After importing the mesh, it was transformed into quads (4/4) or triangles and the vertex density had to be reduced sometimes to less than a quarter of the initial amount.

Finally, the mesh generation and “.obj” export could be executed. Only after these steps were completed could the model be used properly.

2.4. Applying the Photogrammetry Process in Building Restoration

During the annual post-practical work reports, it was noticed that although students apply for office practices, their involvement in the activity undertaken is much lower due to the lack of digital experience. The reports include written and graphical information on their daily practical activities, impressions, and conclusions. Each student presents their personal experience in front of their peers during oral examination sessions that allow for an overview of each generation’s practical activity.

These report presentations have highlighted the fact that students in the first years of practice are not offered the opportunity to be involved in real projects, thus receiving tasks such as hand sketches, mood boards, simple surveys, and rarely small interior designs.

The fact that young students could achieve a skill that is becoming widely spread in the architecture profession and used on many levels, could bring opportunities for better collaboration between local businesses and the academic environment.

This research contained a second practical part that shows the immediate application of digital skills achieved during the workshop, in a professional context. For this second part, the photogrammetry technique was identical, but it was used on a building site and on a scaffold, to document architectural details and create 3D models for future moldings. Because the authors had data regarding the real costs of an ongoing restoration of ornaments (material, workforce, and duration), they could measure better outcomes provided by a different, more digital approach, while using photogrammetry to 3D scan details of the same building.

3. Results

Since the first stage of the research was based on learning a new digital skill, testing, adapting, and experimentation, it was mandatory to follow the progressive achievement of results to understand the level of accessibility this technique has in the academic environment. Once this phase was considered complete, the practical part followed, where feasibility of the same group of students accomplishing real tasks using the same digital technique in a professional environment was of utmost importance. The quality of their work is also proved by successfully printing one of the models, into what can further become a plaster mold for architectural details.

3.1. Progressive Results of the Workshop

The first attempts were made using objects of students’ choice which made the beginning of the exercise more enjoyable. These first attempts were unsuccessful, especially when photographing indoors using a lightbox, thus raising the first questions and problems to be solved. One obvious observation was that the background blended with the surfaces of the object (Figure 2a–c) and the software could not distinguish the object limits, so among the first guesses was the quality of artificial lightning that created high contrasts in the Light Box.

Further on, a team of students tried photographing one object outdoors obtaining successful models (Figure 3) compared to the Light Box model. Discussions were raised and they tried to identify what was the differencing criteria that led to better outdoor scanning. The subjects in question were objects, lightning, and contrast between objects and background or background alone.

After noticing the success of the prior session, the students along with the tutors searched for a method to simulate a dynamic background, similar to the exterior environments but suitable for a Light Box. The condition was that the background rotated along with the object. Consequently, a simple and quick solution was put into place, pasting strips of variously colored paper to the bottom of an object (Figure 4a) and folding them upright so that the camera could capture this created background in each picture. For this experiment, the students used Light Box 3 and brought all the existing lamps to create the most uniform light. This was the most successful method for Light Box photography, as seen in Figure 4b–d, as the result was a proper 3D model without much of the white background blending with the object.

Although most of the Light Box studies were unsuccessful due to the lighting conditions, background, or object type, this initial step had an important scope in two different directions: first, it made students search for various solutions, perfecting their object choice, photography technique, and environmental setup and second, it taught an alternative, more detailed method of 3D scanning aside from the outdoor method.

Students were eager to continue photographing outdoors, as the models were clearly better, so this allowed us to easily advance further with our exercise since the main scope was to scan outdoor architectural elements. This phase involved scanning natural elements (Figure 5) that displayed many surfaces, colors, and texture irregularities, all while benefiting from a cloudy day that gave naturally uniform and filtered light.

3.1.1. Documented Students’ Workflow and Observations

Our request was that students documented their workflow and note and observed each change they made from one iteration to another. This allowed all participants to notice when the technique needed alterations in terms of light, photography angles, motion, and stances, or even when the subject was not suited for the applied method.

The authors found that this simple process of documentation was a valuable tool both in terms of activity management and advancement while teaching freshmen to work in teams, observe, evaluate, and adjust their own ongoing workflow. Thus, students gained independence and confidence in their technique, being able to perform very well during the second week of the workshop that was dedicated to individual study. As seen in

Table 2. Students’ observation worksheet summary.

	Team 1	Team 2	Team 3
Lightbox description	H66 × 53 × 53 cm	H43 × 25 × 53 cm	H72 × 70 × 70 cm
Objects (in chronological order)	Wooden dummies, apples, tree trunks, brick walls, textile, statues, architectural details	Plush toy, tree trunk, two statues, wall, architectural details	Laptop, two plush toys, tree trunks, statue, facade
Attempts	8	10	6
Avg. No. of pictures	103	64	160
Main observations in LightBox conditions	- rotated pictures in Reality Capture - the software doesn’t take all the pictures - objects completely blended in the background	- some pictures are rotated - objects poorly defined but visible	- background blends with an object, only a few details are clear - objects poorly defined
Main observations in outdoor conditions	- good results, clear details	- too much background information - very good results, clear details, and textures	- good results, clear details
Lightning	Artificial, mostly natural

, the first object choices as well as the strategy of building the three Light Boxes, both actions undergone at the beginning of the workshop, were very different from one team to another, while decisions taken later in the workshop tended to become more uniform as students learned from each other’s mistakes and successes.

3.1.2. Scanning Art and Architectural Details: Individual Study

Using all previous experience based on observations, adaptations, and conclusions, the students were able to apply the photogrammetry technique by themselves to document elements specific to their field of study: art sculptures, facade details, and textures. They also understood the importance of accurately scanning these elements to build their own 3D object library for future projects, an important asset for any designer.

After being accustomed to the photogrammetry method during the first week of the on-site workshop under the guidance of the tutors, students were asked to individually search, analyze, and identify optimal objects and surfaces in historical architectural contexts. Since photogrammetry principles imply an attentive selection of the subject, students had to analyze the objects beforehand regarding texture, imperfections, surface typology, glossiness, and volume.

As seen in Figure 6, all the best-photographed objects and surfaces present ununiform surfaces and textures (Figure 6a–d,l) without important protuberances, compact volumetrics (Figure 6e–k), and no glossiness. Students learned to optimize both their photography methods and selection of subjects, compared to the first week of the workshop.

Comparing the results of the scanned models obtained during the first supervised workshop days and the models obtained by students during their individual work, one could easily observe a higher quality and accuracy of models and textures. The authors believe this is due to multiple iterations through a trial-and-error process, that led to better model selection, but the results simultaneously validated the proposed method of documenting architectural and art details through photogrammetry. These types of models, whether they be façade textures or details in stone, plaster, or carved wood, volumetric statues, or decorative objects, were easily accessible for students to observe and their overall structure was appropriate to the photogrammetry method, having the potential to be further applied in restoration projects, as described in the following section.

3.2. Applying Photogrammetry, 3D Modeling, and 3D Printing on Architectural Ornaments: Case Study of a Palace Restoration in Timisoara

3.2.1. Localization and Description of the Lajos (Ludwig) Besch and Károly (Karl) Piffl Palace

The information acquired before and during the workshop offered good knowledge and ideas for future projects, such as the following presented restoration project that started in 2020. The building is a former palace located in a historical area of Timisoara (Figure 7), a city in the western part of Romania.

Timisoara is proud to have an impressive number of historical buildings, over 14,000, most of which are in an advanced process of deterioration after the frequent change of owners and lack of interest in common space maintenance, including facades and ornamentation. This topic was already addressed in one of the authors’ previous articles.

“The nationalization of the properties and interior subdivision of the apartments in historical buildings occurring during the communist regime has put a dangerous mark on the once proud bourgeoisie buildings, including massive structural and aesthetical degradations. After the collapse of communism in December 1989, the population could finally buy these scattered apartments they were living in. Unfortunately, some common spaces remained under nobody’s care, for example, the stairs, the attic, and the courtyards. In the last years, the historical buildings, surroundings, and public spaces entered a long process of restoration. A new life is expected for these areas after several decades of abandonment” [11].

Once the title of European Capital of Culture was earned by the city of Timisoara, a vast process of heritage rehabilitation has begun under the pressure of the city hall; however, it began without feasible co-financing options. Most of the owners who live in the historical areas have managed to independently raise money for rehabilitation but have struggled with high material costs and finding a skilled construction and manufacturing workforce. The challenges of the analyzed project were similar, as the current owners started the restoration process without any funding and help from the city hall, although the city hall still owned 3 apartments of the total of 24.

Regarding the plan configuration, the apartments surround a common interior courtyard, as seen in Figure 8, which is among the few historical courtyards with used green space for leisure and a children’s playground. From a historical point of view, the palace was described in a recent article named “The artists’ house from the number 11” published by the Heritage of Timisoara Project [12]. The title refers to the artists who live and still create art in the building (painters, sculptors, architects, actors, etc.).

Belonging to the 1900s style, a representative of the architecture of Timisoara between 1900–1920, the former palace is part of an impressive architectural heritage, which still marks many of the city’s neighborhoods. Although the differences and demarcations between the various styles belonging to the 1900 style are unclear and fluid, it can be said that Lajos Besch and Károly Piffl Palace belongs to the Art Nouveau/Secession style.

Specific ornaments and characteristics of the main facade are testimonies of the Art Nouveau/Secession movement, a more geometrized stage but with an emphasis on the leitmotif of the sinuous line. Curved lines, medallions, and also the distinguishing wavy attic of the main facade, are all adorned with vegetal motifs, intertwining flower decorations, and the heart symbol (Figure 9). These ornaments can be found even in the smallest details of the carpentry and balcony railings [13].

3.2.2. Restoration Project of the Piffl Palace: Traditional Versus Digital Method

In the initial stages of restoration, the construction team did not know how to correctly collect historical data from the facade, as it was thought that technical drawings would be sufficient (Figure 10). Consequently, during the rehabilitation process, important indications regarding the lines and ornaments were lost. The situation could have been avoided if every portion of the facade had been previously 3D scanned prior to intervention, but a few years ago, 3D scanning was hardly accessible for general use in Romania and at very high costs. However, using the previously presented technique, 3D scanning is within reach, using the most widespread gadget: the personal smartphone.

Considering the challenges that the owners had faced with finding skilled sculptors to restore the Art Nouveau ornaments, the authors have decided to experiment with photogrammetry to find suitable solutions for similar cases. The following section will describe an inexpensive alternative to scan, model, and 3D print ornaments for molding preparation, as compared to the traditional method used by artists or sculptors who shape the ornament by hand [5,14]. The few sculptors that make these interventions nowadays are rare and expensive since their work requires meticulousness, talent, and hard-working conditions on the scaffold.

The traditional method used by sculptors begins with the local restoration of the ornament, or, if it is very degraded, it is removed from the facade and restored in the workshop, as seen in Figure 11. Great care must be taken when trying to detach the piece from the facade, as it is ideal for it to remain unbroken. After deep cleaning, when the finishing layers are removed from the ornament, the following step is the reconstruction of the missing parts. This is where the creative hand and experience of the artist come into play. After completion, they are sanded and primed for hydrophilization and resistance. Those made in the atelier are mounted back on the facade. If the ornaments are repetitive, then after one ornament is reconstructed, molds are made so that the rest can be cast according to the desired number (Figure 12). The procedure is classic and works very well, the only disadvantage being the specialized workforce needed and the increased execution time.

The alternative proposed method is to restore these ornaments by using photogrammetry and an additive manufacturing process after the creation of the digital model. The steps are described as follows:

Step 1—Photogrammetry, 3D scans of the ornament. As seen in Figure 13, the first step is to take gradual pictures of the piece, starting from the upper side and going in a crisscross motion towards the base, with small steps. These sets of pictures are processed through the same phases as previously presented during the workshop until the outcome is an optimal .obj file.

Step 2—3D modeling of the scanned ornament, corrections, and additions. If the ornament is very degraded, it is necessary to correct or reconstruct the mesh using various 3D modeling programs. The software used was 3D Studio Max, as shown in the images below. Throughout this phase, missing areas can be reconstructed, some edges can be rounded, and smoothed, or the unnecessary background area can be removed (Figure 14 and Figure 15). There are many CAD programs, which use different modeling principles, for example, ArchiCAD, Rhino, Solidworks, Autodesk Fusion 360, SketchUp, 3D Studio Max, etc.

Step 3—Saving the model in STL format. After cleaning and remodeling, the digital model is converted into an STL (Stereolithography) file. STL files are used in 3D printing and contain the 3D model to be printed, using triangular facets.

Step 4—STL file import into the program (SLICER) and G-code generation. Once an STL file has been generated, it is imported into a program (slicer) that converts it to G-code. G-code is a programming language used in computer-aided manufacturing (CAM) to control automatic machine tools.

The Slicer software (e.g., Ultimaker Cura or Idea maker) allows editing, and preparing the model, size, and quality of the print, mainly depending on its size (Figure 16).

Step 5—Preparing the 3D printer. This process requires proper setting and control of the 3D printer, cleaning the worktable, and loading the material. A routine check of all the main print settings and the control panel is also required (calibration). Once the equipment is ready, the file can be uploaded for 3D printing.

Step 6—3D printing. 3D printing or additive manufacturing is the process of making three-dimensional solid objects by adding layer after layer [15,16]. 3D printing is especially convenient for custom objects. The 3D printing procedure is almost automatic. Depending on the size of the object, the accuracy set, the materials, and the printer, the procedure can take from a few hours to a few days. In this case, the 3D-printed model took 8 h and 50 min to be completed (Figure 17). It must be checked from time to time for possible errors (due to power cuts, faulty adhesion, and nozzle clogging). We used a very basic, affordable 3D printer that people usually have in their homes or offices. For the final product, it is recommended to use a more efficient printer with higher accuracy for the model to be smooth and have fewer visible construction layers.

Step 7—Final processing. The processing of the object after printing can vary, depending on the printing technology and the materials used. For example, a part printed by SLA must be cured under UV rays while one printed by FDM can be handled immediately. Processing of the final product may include manual or compressed air cleaning, polishing, coloring, and other actions that prepare the product for final use (Figure 18).

Step 8—The molds. Once the printed ornament is prepared, the final step is to create the molds for casting the final plaster parts.

The authors were directly involved in the restoration process of the palace. Because of that, they had clear data about the costs of the material, workforce, and the period of time that the sculptors needed for each ornament while performing the traditional methodology. Conversely, the proposed photogrammetry method that the authors experimented with took place in a much shorter period (of course depending on the users’ experience of managing 3D modeling programs) with a low budget to produce 3D printed models. The best results could be obtained while applying the most appropriate method (traditional versus digital), depending on the stage of degradation of the ornaments, budget, workforce, and owned technology, or experimenting with a combination of using both presented methods, depending on previous analyses of the subject.

4. Discussion

The article proved the functionality of an inexpensive, adaptable, scanning technique that is also easy to learn and quick to apply.

Regarding the educational side, the authors noticed that students who did not interact with digital methods beforehand were easily disappointed when the resulting 3D models were not accurate, but their interest quickly rose back if they managed to obtain good models. This is where the instructors came into play, as they could notice the interest variations and could lead the exercise toward objects or environments that were more likely to lead to accurate models.

However, there are some limitations to the exercise. First, the quality of the models photographed using lightboxes was highly influenced by non-uniform lightning that can only be obtained with multiple filtered light sources; this was a drawback for the “do-it-yourself” method. Second, the various smartphone web cameras also influenced the results. From the educational and specifically observational point of view, tutors must direct students’ attention to observing architectural details when photographing and generating the 3D models, rather than on the method itself. In this case, due to a shortage of time (one week on-site, one week remote), students only managed to get accustomed to the photogrammetry method and workflow, so almost all their attention was concentrated on learning and testing the proper methods rather than observing architectural details.

To ensure the relevance and continuity of the research, the authors propose to continue observing the development of this group of students over time compared to peers with similar backgrounds, in regard to technology manipulation and career integration. This follow-up would prove the efficacy of the presented research over time and observe the benefits for both the students and the professional network.

In order to fully understand the concepts and how to use specialized programs in obtaining 3D models of objects, students must have previously studied subjects such as analog photogrammetry, technical drawing (or CAD), and tools/methods of measuring.

The essential benefits of this type of application are:

• The use of “low-cost” photogrammetric technologies;
• Improvement of classic measurement methods;
• Efficient image processing;
• Obtaining final products in the 3D system;
• The use of new 3D modeling tools and products;
• 3D analysis of heritage objects, such as artifacts.

Terrestrial photogrammetry and stereophotogrammetry methods have the advantage that they can fix constant and temporary deformations with fairly good precision.

5. Conclusions

Photogrammetry has the potential to become a valuable link between the educational and professional fields of architecture and design, with specific applications in restoration.

The objectives of the practical applications were the acquisition of the skills by students aimed at the basics for analog and digital photogrammetry: focal length, orientation elements, interior and exterior of the photogram, coordinate systems used in photogrammetry, and the use of specialized programs for processing.

The main advantages of the use of 3D visualizations of cultural heritage objects in research and communication for design and architecture students are:

• Learning the skills for 3D vectorization of different shapes, surfaces, etc.
• The development of intelligent tools for the acquisition of three-dimensional data relating to heritage objects using the technology available at low costs.

This article mainly presents some case studies, proving that this method is suitable for students and professionals, with all ideas being argued from a scientific point of view.

The summer workshop is not the only case study revealed here, but it could be relevant and it was an excellent platform to verify the efficiency of photogrammetry-related issues and their acceptance on both the students’ part as well as from the academic staff or professionals focused on heritage conservation. This is why we described it in detail as a research and didactic instrument.

Image-based 3D modeling techniques demonstrate the usefulness of digital photogrammetry in modeling and 3D visualization with the precision of real objects that present regular geometric shapes (monuments, buildings, etc.). The precision with which the 3D models are obtained (under a pixel) corresponds to reconstruction applications in the field of historical heritage conservation; thus, photogrammetry constitutes the best alternative to classical measurement techniques.

However, the quality of such an open exercise stands in the development of problem-solving skills in the students, something that is rarely practiced in Romanian teaching and is a valuable career asset. Similar thinking strategies must be applied on-site, as modern-day restoration proves to involve a necessary amount of creativity and digital skill [17].

The method applied in the case study project shows an alternative to traditional ornament manufacturing techniques that could become widespread even in lower-income areas [18]. For further experiments to complete the process, the next challenge is to directly print the 3D ornaments with construction materials, for example, plaster [19]. The next step would involve accessing a grant to financially help us in this direction.

This article complements existing modern techniques of research, analysis, and heritage documentation, among different ones that are presented and developed in various articles that have been previously published [20,21]. Considering the vast number of heritage buildings around the world that need urgent documentation and interventions [22], photogrammetry is a suitable tool for documenting details as soon as possible.

It is an affordable and quick technique for the digital preservation of the built environment and its heritage details.