A 3D quick sketch algorithm in virtual reality for concept design in architectural studio

Introduction

Even though the integration of architecture and design fields with digital technologies occurred in the early 1980s, the computer was viewed and employed primarily as a drawing tool for decades. ¹ Due to the development of computer programs and related technologies in the 2000s, it is evident that computer technologies have begun to play a role in the design process. In this way, the computer has become a tool for designers at every stage, from design to presentation to marketing. In addition, the capabilities provided by computers and related technologies (3D printers, laser cutters, CNC machines, etc.) made it possible to draw and manufacture amorphous geometric forms that were previously difficult for the designers to imagine.

Design is a creative process, and during the preliminary design phase, designers wish to freely express their ideas in three dimensions. Sketching plays a crucial role in the design process as it provides architects and designers with a means to articulate and refine their initial ideas. The inherent value of hand-drawn designs is not derived from their precision, but rather from their capacity to swiftly and fluidly encapsulate the essence of an idea. ² One of the reasons why these designs frequently avoid direct translation into digital models is that precision is not their primary concern. On the contrary, sketches offer a conceptual void that encourages exploration and interpretation while serving as an interface to digital formats. The interpretive phase may involve the utilization of intermediary representations, including exploratory drawings and orthographic projections, to condense and convert the fluidity of a sketch into the methodical language of digital tools. ³ This recognizes that the digital sketching of a notion is not a final destination, but rather a progression in the ongoing discourse between the conceptual and reality.

Edwards ⁴ summarizes the functions and significance of sketching as follows: “The act of drawing is an important starting point for the intellectual process we call ‘design’. To be able to draw a chair or a building is a prerequisite for anyone wishing to design such things. Drawing has two functions for the designer – it allows her or him to record and to analyze existing examples, and the sketch provides the medium with which to test the appearance of some imagined object.”

Suwa and Tversky ⁵ also argue that sketching enables designers to see unexpected relations and features that enables refining and revising ideas, they call this process as having a conversation with one’s self. Goldschmidt ⁶ similarly names sketches as self-generated displays and argues that sketching allows designer to review the whole history of design activities in a given session concurrently.

Sketching as a rough and quickly made drawing plays a vital role in early stage of designing and serves as a thinking tool for designers and facilitates problem solving. Therefore, there is certainly a strong relation between sketching and design creativity.

Although there is a thought that digital systems can be used in early design stage as well as detailed design stages but existing digital systems do not still support the early stages of conceptual design process thoroughly.

At this point, it is possible to say that the boundaries of the conventional sketches framed by a two-dimensional paper plane may be insufficient for expressing complex geometries.^7,8 The searches, that are called algorithmic forms by Terzidis ⁹ and animative forms by Lynn, ¹⁰ hypersurfaces by Goulthrope, ¹¹ tophological geometries by Cache, ¹² liquid architecture by Novak, ¹³ or parametric architecture by Kolarevic ¹⁴ ; refer to forms with flexible surface geometry in which curvilinearity is at the forefront rather than orthogonal lines. For these complex forms, tools are needed that can bend the lines and control the degree of curvature. Terzidis ⁹ and Tedeschi ¹⁵ stated that in CAD software, designers’ control over the design is restricted by ready-made buttons with defined functions and designers cannot be free on the screen. They state that direct commands (scripting) and algorithms that enable intervention in the digital screen are tools to overcome these limits.

The use of existing CAD software to create three-dimensional conceptual designs is time-consuming and can become a significant burden, particularly when considering that design proposals are frequently modified at the beginning. Furthermore, three-dimensional modeling with CAD software necessitates design expertise and precise (in terms of shape, size, etc.) information about the final product. However, in the conceptual design phase, many elements often lack clarity and fail to be expressed concretely. This situation may result in the inability to adequately express ideas in the minds of especially young designer candidates, at the beginning of their education life. On the other hand in the initial phases of design, students turn to traditional CAD systems, which are inadequate as they restrict them to predetermined parameters and do not possess the intrinsic adaptability that is characteristic of manual sketching. Accordingly, using the capabilities of today’s CAD programs in architectural design education, it is extremely difficult to quickly create 3D concept shapes based on the abstract approaches in the mind, during the early stages of design and to revise the designs in response to the critics from the professors.

With the rapid advancement of computer hardware, especially processors (CPU) and graphic processors (GPU), CAD contributes to 3D modeling of complex forms. Nonetheless, during the production of different alternatives at the beginning of the design process, the designer’s interaction with the model is limited to a mouse and a monitor, causing uncertainty in terms of scale, real-size perception, and product-space relationship. Therefore, the keyboard- and mouse-based interfaces make it difficult to use existing systems for rapid three-dimensional design generation and exploration. ¹⁶

In this context, this study offers a practical and integrated tool with 3D modeling systems that will allow students to experiment with more alternatives during the concept development phase. As the name suggests, sketching suggests the ability to draw freely and quickly with a single tool (pencil, brush, finger, or computer mouse). When the designer decides to sketch, he needs a seamless and fluid space without limitations. Therefore, the purpose of our medium is to be a facilitator in this sense and not to replace traditional sketching, but to be a tool that supports it. By using this 3D quick sketching algorithm (3DQSA), students could interact with the three-dimensional model by entering a Virtual Reality (VR) environment. The main goal is to preserve the fluidity and interpretive capacity of hand-drawn drawings and increase the accuracy of the transition to 3D so as through the use of VR and also to augment the spatial reasoning that is fundamental to design thinking by incorporating VR into the architectural sketching procedure.

The sketching process is a step for testing and questioning various alternatives, especially in design education. The fact that the student bypasses the 2D sketching process in favor of computer modeling directly is one of the most fundamental difficulties encountered in the educational system. Conventionally, the design in mind is transferred to paper with 2D handmade sketches; various options are considered; and then 3D modeling is performed using computer programs by selecting one of the possibilities. At this point, students often limit the number of 2D sketch suggestions due to the perceived hardship of producing many alternatives by hand. Thus, the student bypasses the two-dimensional early sketching process and directly engages in computer modeling. On the other hand, because the 3D modeling process is a time-consuming activity for many students, typical results are the selection of one or two of the possible 2D sketches and their transformation into models, while other possibilities are eliminated prior to their visual realization. The inefficiency of the final product, its crudeness, and the occurrence of numerous errors are the result of being skipped to the modeling phase without questioning the appropriate alternatives with the sketch.

The main purpose of this study is to provide an alternative solution for the need for rapid and practical 3D sketching and model development at the concept design stage in architectural education by utilizing VR technology and to investigate the potential of VR to supplement conventional tools to augment spatial comprehension. It enables the student to experience the design proposals conceived in his mind as a whole in a 3D environment with desired scale. Thanks to VR technology, the student is able to describe their preliminary sketches with rapid hand movements, observe their gestures into a three-dimensional model, and perceive them in their actual scales and dimensions. Thus, it is thought that, VR will provide an immersive enhancement to conventional abilities, allowing architects and students to interact more intuitively and physically with their designs. By means of VR’s immediacy, the intricacies of sketching can be converted into a dynamic and interactive three-dimensional model, enabling an investigation of form and space that surpasses the constraints imposed by the paper plane. The research believes that VR should not supplant the essential abilities of sketching, but will enrich the designer’s repertoire and enhance creative processes by offering new equipment of communication and perception. Accordingly, 3DQSA is developed as a supplementary instrument rather than a substitute.

The fundamental aims of the study are:

- To allow the student to interact with a 3D interface instead of a 2D interface during the design that will be manufactured in 3D, to enable the student to change the scale, thereby experiencing the perception of real scale and dimension.

Methodology

The research methodology utilized in this study comprises several stages. It begins with a review of the literature to lay the groundwork for a comprehension of VR and VR sketching as they pertain to architecture. The objective of this preliminary stage is to conduct a comprehensive review of current research in order to identify fundamental principles, obstacles, and developments that pertain to the incorporation of VR technologies into architectural design procedures. The research follows to the real world by doing a pilot modeling study in the parametric design environment of the Grasshopper plug-in within Rhinoceros 3D software.

After the pilot study, the methodology moves on to developing and improving the algorithm based on what was learned. This pivotal phase entails the improvement of current algorithms, or the development of innovative ones customized to meet the particular demands of parametric design in VR. The process’s iterative nature facilitates algorithm optimization, thereby utilizing algorithms’ capability to adjust to the intricacies of architectural design and VR interaction.

After the development of the algorithm, the research proceeds to a phase of practical testing wherein students participate in an architectural design studio. The objective of this collaborative segment is to evaluate the efficacy and practicality of the enhanced algorithms within an educational environment.

The concluding stages of the methodology consist of an analysis of the results, establishing correlations between the literature review, pilot study, algorithm development, and student survey. The discussion aims to consolidate the knowledge acquired during the research and contribute to the wider dialogue on incorporating VR and parametric design in architectural education. In its final section, the article provides a synopsis of discoveries and possible directions for further investigation at this convergence of technology and architectural design pedagogy (Figure 1).OPEN IN VIEWER

Literature on 3D modeling technologies and VR in architecture

In the early stages of computer-aided design, architects realized that, despite the fact that the computer screen is two-dimensional, all types of three-dimensional forms can be represented in the computer environment. The doctoral dissertation “Sketchpad” by computer scientist Ivan Sutherland was the first instance of digital sketching in 1964. ¹⁷ Users can draw vector shapes by moving a pen equipped with a sensor across a monitor.

Virtual spaces and architecture are being reevaluated using computer-based advanced motion tools in the present day. The development of graphics processors reveals the use of computers in architecture. First, architects determined that they could create architectural technical drawings in two dimensions using computers. However, it took a little longer for the ability to create three-dimensional designs on computers to become prevalent and enter the market. With the advent of graphics processors and the ability to create three-dimensional models, visualization (rendering) techniques have emerged as a new architectural representation. Simultaneously, the development of processors and the increase in calculation speed made photo-realistic representations possible by paving the way for light simulations.

The investigation of three-dimensional modeling transcends architectural education and has emerged as a pivotal component in fields that require the fabrication of complex geometrical forms. Digital sculpting in the fields of concept art and game design has been significantly transformed by the introduction of tools like Blender, ¹⁸ ZBrush^19,20 and Adobe Substance 3D Painter. ²¹ These applications enable artists to effortlessly generate intricate textures and forms by utilizing a stylus and tablet, thereby replicating the physical sculpting experience.^22,23 The progression of virtual sculpting is further exemplified by the introduction of Kodon, ²⁴ which offers a screen-based interface that connects two-dimensional perception with three-dimensional execution. Moreover, for the purpose of visualizing complex designs from multiple vantage points points, Blender’s Grease Pencil tool exemplifies the inventive translation of two-dimensional references into three-dimensional space by enabling artists to draw directly in a three-dimensional viewport and manipulate lines on planes in a variety of orientations. ²⁵ These technologies illuminate the convergence of artistic autonomy and digital creativity, thereby enhancing our comprehension of the potential applications of similar principles in architectural visualization.

Sandbox games, which are widely prevalent in the realm of interactive media, exemplify the dynamic way in which users and digital tools collaborate to construct three-dimensional models. Games such as ²⁶ and ²⁷ provide users with the opportunity to interact directly with the game world, constructing complex structures and landscapes via user-friendly interfaces that promote exploration and experimentation in design. This genre demonstrates how gamified interfaces have the capacity to democratize the process of 3D modeling, rendering it accessible and captivating to a more extensive demographic.

The sandbox games’ underlying principles find parallels in the progressions of Extended Reality (XR) and VR, wherein the indistinct separation between the real and virtual worlds provides enhanced levels of engagement and interaction. ²⁸ Users have the ability to not only create and modify digital environments in XR and VR, but also to inhabit and experience them. This expands the scope of architectural design by introducing novel opportunities that are shaped by and contribute to these interactive paradigms.

Interpretation of Mixed Reality (MR) is always dependent on context. ²⁹ Early definitions of MR were as follows: augmented reality (AR) and augmented virtuality (AV) were considered to be special cases of MR. AR projects virtual objects onto the physical environment, whereas AV incorporates physical objects into a virtual environment. ²⁸ According to some recent studies, MR systems are defined as blending physical objects in at least one physical environment with virtual objects in at least one virtual environment and operating interactively and spatially to map physical and virtual objects to one another. ³⁰

VR is conceptually divided into two according to the state of technology in the late 1990s and early 2000s. The first is immersive VR. This definition aligns precisely with the concept of VR utilized today. In non-immersive VR, virtuality encompasses the visual perception of the subject with a matching headset or a visual projected into the space. Dani & Gadh provide non-immersive VR through the use of stereo glasses and a compatible screen. The screen size limits the perspective of the experiencer. ³¹

Researchers have been attempting to apply methods for producing sketches in the VR environment since the mid-90s. The devices and interaction tools utilized in VR research are the most influential determinants of the user experience and the efficiency of the working environment. For interaction, mouse ³² and wands ³³ are typically used in the early stages of VR research, along with joysticks. In the majority of studies conducted during this time frame, non-immersive systems devoid of Head Mounted Display (HMD) were utilized.^34,35 In the majority of the studies, anaglyph stereo glasses were used, but with the advent of shutter glasses, the image quality improved, and a variety of interaction devices, such as sensitive gloves,^36,37 were developed.

With the increasing prevalence of tablet technology, graphic tablets and tablet computers have become interaction tools.^38,39 Thanks to recently developed sensors such as Leap Motion, it is also possible to interact with hand and finger movements.^40,41

Research conducted by Waller et al. ⁴² on the Hive as an immersive virtual environment for spatial cognition research offers insights into human interaction within expansive physical environments in VR. These findings have the potential to impact the processes involved in architectural design. ⁴² In addition, Cha et al.'s investigation into spatial perceptions in Hive highlights the capacity of immersive virtual environments to provide realistic representations, thereby facilitating end-user engagement in architectural design reviews. ⁴³

In recent years, there has been a significant amount of research conducted on the function of immersive environments in architectural design. The work of Tomas Dorta and his associates has provided significant insights into this field, which is worth noting. Dorta, Kinayoglu, and Hoffmann examine the Hyve-3D system, which incorporates conventional sketching methods into a 3D VR environment. This system facilitates a more intuitive design process by enabling architects to seamlessly transition from 2D outlines to 3D models. ⁴⁴

Alongside LaLande, Dorta’s prior research examines the more extensive implications of VR on the architectural design process. Their early research emphasizes the potential of VR to improve the overall design outcome by fostering spatial comprehension and collaboration among design team members. ⁴⁵ Underscoring the transformative potential of VR in architectural design, these seminal works pointed out the integration of traditional and digital design methods. Dorta and Kinayoglu ⁴⁶ argued that the current digital paradigm of the design studios, arbitrarily combining conventional computing and traditional representations, does not support design discussions, and is based on the pictorial-frame of 2D/3D renderings requiring scaled representations and they proposed to fully hybrid (analog/digital) studio concept. Abdelmounema and Dorta, ⁴⁷ introduced an approach that rough freehand 3D sketches were 3D printed as a new representation mode during ideation. They consider that especially with VR systems like Hyve-3D can be a useful tool to go from 3D sketched ideas to the fabrication of these ideas as physical representations.

With these advancements, research has begun on space perception, ⁴⁸ spatial perception, ⁴⁹ cognitive activities, ⁵⁰ and emotional state studies. ⁵¹ Furthermore, it is known that design studies conducted in spaces created by the reproduction of the natural environment in a virtual environment, such as a forest, boost creativity and motivation. ⁵² Furthermore, Seybold and Mantwill found that sketch production in the VR environment is significantly faster compared to traditional sketches. ¹⁶

Model retrieval is another field of study that employs 3D sketch creation while modeling in a VR environment. ⁵³ Students define the objects they wish to call from the software by creating a rough sketch of the object. The system recognizes the object group the student wishes to acquire and displays the corresponding menu on the interface.^54,55

In the concept phase of design, sketch modeling studies in the VR environment are typically used, as in this study.^39,56 Gravity Sketch is the most widely used, the most comprehensive, the most designable form variation, and the software that can switch between simple and advanced interfaces based on the user’s profile, according to Ekströmer. ⁵⁷

Rahimian et al. ⁵⁸ conducted a feasibility study regarding the replacement of non-intuitive CAD tools with VR 3D sketching interfaces. ⁵⁰ They discovered that novice designers’ cognitive and collaborative design activities are enhanced by VR 3D sketching, indicating a substantial potential for digitalizing the conceptual architectural design phase.

Rahimian and Ibrahim ⁵⁰ identify the effect of integrating the thoughts and actions of inexperienced designers into a VR 3D sketching in 3D interface through the use of a haptic-based interface, which enhances their design creativity. Their research indicates that the utilization of these interfaces can improve cognitive and collaborative tasks, resulting in greater involvement in the design process. ⁵⁸

Yang and Lee examine the cognitive effects of VR sketching in the phase of ideation. VR sketching was discovered to expand the solution space, facilitate the transformation of ideas, and encourage a holistic design approach. These results offer significant insights into the cognitive function of digital design tools and should be considered in future design research. ⁵⁹

With the introduction of Mobi3DSketch, Kwan and Fu tackle the difficulties associated with 3D sketching in mobile augmented reality. The authors illustrate the potential of mobile AR in architectural sketching by showcasing how the integration of multiple input sources, along with 3D snapping and surface proxies, can facilitate the in creation of 3D concept designs. ⁶⁰

Lipson and Shpitalni examine the capacity of CAD systems to comprehend three-dimensional models represented in solitary hand-drawn sketches. This would enable the integration of well-established analysis tools during the initial phases of design, facilitating conceptual analysis via the medium of sketching. ⁶¹

Incorporating VR and algorithmic sketching support into the architectural design process, particularly during the conceptual phase, may significantly improve designers’ spatial cognition, collaborative abilities, and overall creativity, according to the findings of these studies.

Current state in VR sketching

Creation and visualization of the 3D design concept are crucial components of the design. The user interface of widely used CAD software remains largely unchanged, and rapid 3D has become one of the bottlenecks in conceptual design. The keyboard and mouse based interface mandates the use of existing CAD systems for rapid generation and exploration of 3D conceptual designs. When students are in a state of creative inspiration, they want to express their design ideas in 3D as quickly and unobtrusively as possible. Additionally, 3D modeling requires comprehensive, concrete, and precise design knowledge.

Existing CAD systems make it difficult for students to rapidly create 3D concept shapes and modify their designs. The majority of research on modeling in the VR environment focuses on modeling diversity and precision in the virtual environment. A current research focuses on the perception of hand and finger movements in the context of virtual pottery. ⁶² Similar to pottery, hand shapes manipulate the radial form model in related studies. ⁶³ In another example, although Fuge and his colleagues’ study of model creation with finger movements ⁶⁴ aims to create a concept, the modeling process is based on the exact solid model production method. Due to the single camera angle and desktop working area, there are also limitations on movement and gestures.

3D quick sketch algorithm through VR in a parametric design environment

We conducted a search for VR-compatible software in the Rhinoceros-Grasshopper environment, where we developed the study and completed the necessary coding. The search for HMD-capable plugins is prioritized in Rhinoceros software, which serves as the operating system for the Grasshopper plug-in. The MindeskVR add-on, which runs within the Rhinoceros program, has been installed for use in preliminary curve strokes. After the curves are drawn in the VR, the algorithm runs and computes the form, allowing the VR experiencer to view the final model. This procedure takes approximately three to 10 seconds, according to the size and complexity of the sketch.

After drawing a three-dimensional polyline in the VR environment using the MindeskVR plugin for the Rhinoceros interface, the resulting polyline was defined in the Grasshopper interface. An algorithm that averages the sketch strokes of the user processed the main concept of the proposed model. An experiment modeled action using a 16 × 16 × 16 three-dimensional grid. It was observed that the obtained geometry consists of an excessive number of untreated cubic geometries. We attempted to obtain a model with greater detail by increasing the number of grid cells. When a grid consisting of 32 × 32 × 32 cells is utilized, it was observed that the formed geometry is significantly more detailed than the drawn polyline and that the source lines provide a more precise description of the volume (Figure 2).OPEN IN VIEWER

During the development of the algorithm, a three-dimensional curve was manually defined in the Rhinoceros software’s user interface. Subsequently, it is anticipated that the algorithm will produce a shape resembling a quarter-sphere and that the evaluation will be conducted in a healthy manner. In the initial phase of the algorithm, the CenterBox command created a cube. The 3dArray command was then used to replicate this cube. We obtained the centers of the replicated cubes by using the center output of the Area command. Then, the curve component of the Rhinoceros software was introduced to the three-dimensional line drawn in the VR environment. Using the two aforementioned data, the CurveClosestPoint command determined the proximity of the cubes to the three-dimensional curve in the three-dimensional grid. The Smaller and then Cull Pattern commands separate the cubes with 15 units close to the curve in the three-dimensional grid, revealing them. The same method was applied to cubes with a distance of 10 units. Using the CreateSet command, we created a cluster for both groups and used the SetDifference component to extract the units from the segments of these two clusters. The ClosestPoints component transferred the data from the intersection set and points less than 10 units in size. The Mass-Addition command compiled the distances between the closest points. The cull command is used to separate data that is less than a certain threshold and comes from this command. Finally, all data were merged using the merge command, and the algorithm was finished (Figure 3).OPEN IN VIEWER

Testing the 3DQSA in a workshop

A workshop on “3D sketching in VR” was organized to test the 3DQSA within the scope of MIM1502 Introduction to Architectural Design, in the third phase of the research. Students were required to design a sleeping capsule for a space shuttle that would travel to the moon, according to the course’s setup. The course distributed the necessary anthropometric information for a space capsule to the students as leaflets. The students were assigned a 1/10 drawing and a 1/20 model of the capsule they had designed as homework.

The projects of three students containing the most intricate and curvilinear geometries were chosen based on an examination of their pre-course submissions. Beginning of the workshop, students were briefly explained how the system works, its advantages and how to use the tool. Then, the students used the VR system installed in the studio to create three-dimensional drawings of their designs. Using the hand controls of the VR system, the students were able to freely imagine the form without the intervention of an operator during the sketching process (Figure 4). Emphasizing the use of a three-dimensional curve, they form the entire design.OPEN IN VIEWER

Instructed the subjects to model in the VR environment without scaling and to draw the 3D curve at 1/1 scale. As a result, we omitted the conventional early design phase’s manual sketches and executed a 1/1 design action on a real design scale. The workshop-obtained three-dimensional curves are the fundamental input data for the modeling algorithm (Figure 5).OPEN IN VIEWER

The masses were created using the “metaball” and “voxel smoothing” algorithms, as well as the 3D quadrat analysis algorithm envisioned at the outset of the project. Many variables exist within algorithms. The most significant of these variables are those that regulate model development. Upon examining the algorithms, we have determined that voxel smoothing is the most efficient in terms of processing power savings. However, quadrat analysis produces the closest model to the three-dimensional curves drawn by the subjects. As the length and complexity of the curve modeled in the algorithm’s sketch increase, the algorithm’s processing time increases.

While keeping the algorithm’s processing time ranged between three and 10 seconds, in the subsequent step, the polygon count of the mesh generated by the algorithm was decreased, and the mesh was simplified and made ready for 3D printing (Figure 6).OPEN IN VIEWER

The capacity to alter curves in Rhinoceros serves as a strong basis for iterative design, enabling the precise adjustment of spatial shapes and geometries. By utilizing the parametric capabilities of Grasshopper, our algorithm is able to dynamically adjust the 3D model in response to these modifications. The seamless integration of Rhinoceros and Grasshopper allows for automatic updates to the evolving design based on changes made to the initial sketches. This facilitates a fluid and flexible design process that is both intuitive and responsive to the students’ vision.

Student feedback on VR sketching

A survey was administered to the students who participated in the workshops, allowing them to rate their experiences on a five-point Likert scale, that five was indicative of the most advantageous feature and 1 was the least advantageous. A total of 10 students who used the 3DQSA answered the survey. The results of the survey administered to architecture students concerning their utilization of VR sketching tools provided insights into the manner in which this technology is incorporated into the design process.

The contribution of sketching with VR to 3D thinking was found to be extremely beneficial with a score of 4.6. This implies that VR technology has a substantial impact on improving spatial cognition and is considered a valuable instrument in the early stages of architectural design.

The average rating for gaining new perspectives in design generating process through VR sketching was 3.3, highlighting the capacity of VR to provide innovative insights for design projects. In a similar vein, its benefits in understanding true spatial scale through VR drawing received an average score of 4.6. This underscores the practicality of the technology in enabling users to visualize and manipulate designs at their accurate proportions.

While the usability of VR sketching in the early stages of design was 4.2 points; the average score regarding the practicality of obtaining models with VR sketching was 3.5. This signifies a favorable reception, albeit diminished in comparison to other questions, potentially attributable to a period of adjustment or usability difficulties that could be resolved in order to enhance the user experience. In addition, the benefits of being able to easily change scale while doing tactile modeling were emphasized by all participants.

The fact that all questions received over 3.0 points indicates that the benefits of VR sketching in architectural education are widely recognized. The findings of the survey reveal that the majority of students agree that VR sketching is a useful and convenient tool that promotes a more profound involvement with design concepts. However, the results also indicate that the VR sketching workflow should be improved in terms of usability and adaptability.

Through the integration of these survey results, technology developers and educators can further enhance VR tools to more effectively cater to the requirements of architecture students and professionals. This would effectively streamline the process from conceptualization to the implementation of practical designs.

Results and discussion

Upon examination of the project’s outputs, it was determined that the models contained sufficient detail for the concept stage. Holes are observed in the model where the density of the three-dimensional curves is diluted in the sketch. At this point, we made adjustments to the algorithm and attempted to increase the shell thickness as a solution. Excessive increases in the shell thickness of this algorithm variable result in a decrease in the model’s level of detail.

Although the student can see the lines she or he draws interactively during the sketching process, the most significant drawback is that the model created simultaneously during VR sketching cannot be viewed. This is because the developed algorithm requires processing power and time. Future students will be able to view the model of the sketch in real time thanks to the increase in computer processing power.

The study examines the distinct capabilities of VR in order to improve spatial perception and enable a more interactive investigation of architectural shapes. Through the incorporation of VR into the design process, we investigate its capacity to stimulate understandings and enable designers to engage with and inhabit their concepts in ways that were unattainable using traditional approaches exclusively. Our research presents VR not as a substitute for conventional design approaches, but rather as a supplementary tool that enhances and broadens the creative prowess of the designer. Our argument is that VR has the potential to connect theoretical concepts with tangible implementation, enabling designers to navigate these realms with enhanced agility and understanding.

Conclusion and further studies

At the concept stage, it appears that the VR sketching method is a very effective way for students to rapidly visualize their designs in their minds. Obtaining a physical model by combining this method with the 3D printing technique enables the practical production of different concept-stage alternatives.

Comparing the time students spent in the workshop to the resulting three-dimensional model demonstrates that the algorithm is effective. Using conventional CAD software, it may take hours to develop a concept model with the desired level of detail. The method developed in this study reduces the time to between 5 and 10 min, depending on the student’s ability and the complexity of the design. In addition, as mentioned at the beginning of the study, the students create the model based on their subjective experience of the space by sketching in the VR environment at a desired 3D scale. Keyboard, mouse, and screen interaction in a CAD environment cannot realize this experience. Also, the ability to adjust and observe a model’s scale in real-time within a VR environment represents a significant educational advantage. It allows students to understand spatial dimensions and relationships more deeply, enhancing their grasp of architectural scales.

As a result of simplifying the equipment used in future studies, the sketching method will become much simpler. With more mobile systems that do not require external sensor setup, the VR sketching method can become considerably more applicable. The VR sketching technique utilized in this research is applicable to the AR environment as well. Since the commands used in the algorithm model are in accordance with the data from the hand controllers, whether the medium is VR or AR has no effect on the outcome. Students would be anticipated to employ this method more frequently due to the availability of VR headsets and even AR glasses or lenses in the future.

As a result, this study aims to develop a tool that will enable students to make quick sketches in the early stages of design, especially in order to better understand complex geometries. The limitations of this study are that this tool is still at an initial level and was tested with a narrow group. To understand and analyze the effects of VR drawing on design quality, a comprehensive research over a much longer period is needed. However in future studies, there is a need to develop the tool so that the designer can see their sketch made with the VR tool simultaneously on the model, and also to turn it into an cross platform application.