2D Sketches Research Articles

Comparison of actual and virtual pressure of athletic clothing in active poses

PurposeThe circular design process in contemporary fashion design, from two-dimensional (2D) sketching and pattern making to three-dimensional (3D) prototypes, can be facilitated by virtual prototyping. Virtual pressure representations on avatars provide visual and quantitative information regarding garment fit and comfort, which are particularly important for active wear. The purpose of this study is to investigate the benefits of using avatars in active poses from 3D body scans and the use of digital 3D tools for the design process and the prediction of fit of active wear.Design/methodology/approachThis research initially explores virtual fit of cycling wear in active poses and compares the actual pressure values from humans with virtual pressure maps on custom avatars made from body scans in cycling poses across a range of sizes.FindingsSimilar fit results were achieved visually in both the standing and cycling poses. However, the comparisons showed no correlation between the actual and virtual pressure data. Of the 32 cases representing different combinations of the parameters of this research (four sizes, two garment types, four active poses), the differences were significant. The results suggest that, rather than providing a direct correlation with pressure values on the body, the main value of avatar data is in providing comparative visual support for fit evaluation.Originality/valueThe approach taken in this research, which considers the active pose and the size range, potentially contributes to the improvement of virtual fit technology, and its more effective use in apparel product development and fit evaluation.

THE DEVELOPMENT OF 3D ANIMATION CARTOON ENTITLED MACHINE LEARNING

The research was conducted as follows: 1) to design and develop 3D animation cartoon titled “Machine learning” 2) to evaluate the efficiency of the 3D animation cartoon titled “Machine learning,” 3) to Compare learner learning achievement, and 4) to study the learner’s satisfaction towards the 3D animation cartoon titled “Machine learning,” using six media efficiency experts and a sample group of 25 undergraduate students from Rajamangala University of Technology Suvarnabhumi, Suphanburi Campus, selected through a purposive sampling technique. The implementation of this research project was developed according to the animation creation process, which consisted of: 1) pre-production, 2) production, and 3) post-production. The research instruments employed in this project included estimation scales divided into five levels and a pre-test and posttest, administered before and after media viewing. Statistics used in t-test dependent research to compare achievement. The research results revealed that 1) the efficiency of the 3D animation cartoon titled “Machine learning” was found to be at the highest level with an average score of 4.57 and a standard deviation of .27, 2) that comparing knowledge scores before and after viewing the 3D animation cartoon titled "Machine learning," the average score before viewing was 5.36, and the average score after viewing was 13.12. When comparing the scores before and after viewing the media, it can be concluded that the test scores of the students after viewing the media were higher than those before viewing the media, and this difference was deemed statistically significant at the .05 level.

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

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

OpenECAD: An efficient visual language model for editable 3D-CAD design

Computer-aided design (CAD) tools are utilized in the manufacturing industry for modeling everything from cups to spacecraft. These programs are complex to use and typically require years of training and experience to master. Structured and well-constrained 2D sketches and 3D constructions are crucial components of CAD modeling. A well-executed CAD model can be seamlessly integrated into the manufacturing process, thereby enhancing production efficiency. Deep generative models of 3D shapes and 3D object reconstruction models have garnered significant research interest. However, most of these models produce discrete forms of 3D objects that are not editable. Moreover, the few models based on CAD operations often have substantial input restrictions. In this work, we fine-tuned pre-trained models to create OpenECAD models (0.55B, 0.89B, 2.4B and 3.1B), leveraging the visual, logical, coding, and general capabilities of visual language models. OpenECAD models can process images of 3D designs as input and generate highly structured 2D sketches and 3D construction commands, ensuring that the designs are editable. These outputs can be directly used with existing CAD tools’ APIs to generate project files. To train our network, we created a series of OpenECAD datasets. These datasets are derived from existing public CAD datasets, adjusted and augmented to meet the specific requirements of vision language model (VLM) training. Additionally, we have introduced an approach that utilizes dependency relationships to define and generate sketches, further enriching the content and functionality of the datasets.

Toward More Comprehensive Evaluations of 3D Immersive Sketching, Drawing, and Painting.

To understand current practice and explore the potential for more comprehensive evaluations of 3D immersive sketching, drawing, and painting, we present a survey of evaluation methodologies used in existing 3D sketching research, a breakdown and discussion of important phases (sub-tasks) in the 3D sketching process, and a framework that suggests how these factors can inform evaluation strategies in future 3D sketching research. Existing evaluations identified in the survey are organized and discussed within three high-level categories: 1) evaluating the 3D sketching activity, 2) evaluating 3D sketching tools, and 3) evaluating 3D sketching artifacts. The new framework suggests targeting evaluations to one or more of these categories and identifying relevant user populations. In addition, building upon the discussion of the different phases of the 3D sketching process, the framework suggests to evaluate relevant sketching tasks, which may range from low-level perception and hand movements to high-level conceptual design. Finally, we discuss limitations and challenges that arise when evaluating 3D sketching, including a lack of standardization of evaluation methods and multiple, potentially conflicting, ways to evaluate the same task and user interface usability; we also identify opportunities for more holistic evaluations. We hope the results can contribute to accelerating research in this domain and, ultimately, broad adoption of immersive sketching systems.

SketchMetaFace: A Learning-Based Sketching Interface for High-Fidelity 3D Character Face Modeling.

Modeling 3D avatars benefits various application scenarios such as AR/VR, gaming, and filming. Character faces contribute significant diversity and vividity as a vital component of avatars. However, building 3D character face models usually requires a heavy workload with commercial tools, even for experienced artists. Various existing sketch-based tools fail to support amateurs in modeling diverse facial shapes and rich geometric details. In this article, we present SketchMetaFace - a sketching system targeting amateur users to model high-fidelity 3D faces in minutes. We carefully design both the user interface and the underlying algorithm. First, curvature-aware strokes are adopted to better support the controllability of carving facial details. Second, considering the key problem of mapping a 2D sketch map to a 3D model, we develop a novel learning-based method termed "Implicit and Depth Guided Mesh Modeling" (IDGMM). It fuses the advantages of mesh, implicit, and depth representations to achieve high-quality results with high efficiency. In addition, to further support usability, we present a coarse-to-fine 2D sketching interface design and a data-driven stroke suggestion tool. User studies demonstrate the superiority of our system over existing modeling tools in terms of the ease to use and visual quality of results. Experimental analyses also show that IDGMM reaches a better trade-off between accuracy and efficiency.

3Doodle: Compact Abstraction of Objects with 3D Strokes

While free-hand sketching has long served as an efficient representation to convey characteristics of an object, they are often subjective, deviating significantly from realistic representations. Moreover, sketches are not consistent for arbitrary viewpoints, making it hard to catch 3D shapes. We propose 3Dooole, generating descriptive and view-consistent sketch images given multi-view images of the target object. Our method is based on the idea that a set of 3D strokes can efficiently represent 3D structural information and render view-consistent 2D sketches. We express 2D sketches as a union of view-independent and view-dependent components. 3D cubic Bézier curves indicate view-independent 3D feature lines, while contours of superquadrics express a smooth outline of the volume of varying viewpoints. Our pipeline directly optimizes the parameters of 3D stroke primitives to minimize perceptual losses in a fully differentiable manner. The resulting sparse set of 3D strokes can be rendered as abstract sketches containing essential 3D characteristic shapes of various objects. We demonstrate that 3Doodle can faithfully express concepts of the original images compared with recent sketch generation approaches. 1

SketchCleanGAN: A generative network to enhance and correct query sketches for improving 3D CAD model retrieval systems

Given an input query, a search and retrieval system fetches relevant information from a dataset. In the Engineering domain, such a system is beneficial for tasks such as design reuse. A two-dimensional (2D) sketch is more conducive for an end user to give as a query than a three-dimensional (3D) object. Such query sketches, nevertheless, will inevitably contain defects like incomplete lines, mesh lines, overdrawn areas, missing areas, etc. Since a retrieval system’s results are only as good as the query, it is necessary to improve the query sketches.In this paper, the problem of transforming a defective CAD sketch into a defect-free sketch is addressed using Generative Adversarial Networks (GANs), which, to the best of our knowledge, has not been investigated before. We first create a dataset of 534 hand-drawn sketches by tracing the boundaries of images of CAD models. We then pair the corrected sketches with their corresponding defective sketches and use them for training a C-WGAN (Conditional Wasserstein Generative Adversarial Network), called SketchCleanGAN. We model the transformation from defective to defect-free sketch as a factorization of the defective input sketch and then translate it to the space of defect-free sketch. We propose a three-branch strategy to this problem. Ablation studies and comparisons with other state-of-the-art techniques demonstrate the efficacy of the proposed technique. Additionally, we also contribute to a dataset of around 58000 improved sketches using the proposed framework.

3D surgical planning method for lower jaw osteotomies applied to facial feminization surgery

Our proposed method uses a three-dimensional (3D) measurement approach that focuses mainly on the lower jaw from basal, lateral, and frontal views applied to the volumetric skull model derived from a computed tomography (CT) of the head. Likewise, we discuss the geometrical features and clinical considerations involved in the 3D biomodeling of the surgical osteotomy. The workflow that allowed this virtual planning to be developed was composed of medical imaging processing software, data extraction software from images, and statistical software that allows the creation and generation of curve-fitting (nonlinear regression) graphs from data. Thirty-two (32) anatomical points were positioned, sixteen (16) measurements were taken, and two-dimensional (2D) sketches in three views (frontal, lateral, and inferior) were generated to overlap in a 3D environment, which informed the cutting of the desired bone segments. Implementing a nonlinear regression curve-fitting on the contours of the original jaws allowed optimal planning of the osteotomy. Desired cutting shapes were extrapolated for the front view by third-order equations, while for the side and bottom views, log-normal distribution curves and second-order polynomial curves were used, respectively. The reduction in the mandibular volume was between 6.55 and 10.27 %, with two of the most important measurements related to vertical reduction in the lateral views and the difference to determine gonion reduction.

VResin: Externalizing spatial memory into 3D sketch maps

An intuitive way to externalize spatial memory is to sketch it. Compared to traditional paper-based sketches, virtual reality (VR) creates new opportunities to investigate the 3D aspect of spatial memory as it empowers users to express 3D information on a 3D interface directly. The goal of this study is to design a 3D sketch mapping tool for researchers and non-expert users without sketching expertise that enables externalizing memories of spatial information after some 3D-critical tasks. There exist 3D sketching tools using VR, but there are two issues with the current mid-air 3D sketching approach: (1) distortion of sketches due to depth perception errors and (2) increased cognitive and sensorimotor demands due to an increased degree of freedom and absence of physical support. To address these problems, we implemented VResin, a novel sketching interface that synergizes 3D mid-air sketching with 2D surface sketching to scaffold 3D sketching into a layer-by-layer process. An experimental study with 48 participants on multi-layer building scenarios showed that VResin supports users in creating less distorted sketches while maintaining the level of completeness and generalization compared to mid-air sketching in VR. We also demonstrate the potential applications that can benefit from 3D sketch maps and the suitability of VResin for a variety of building shapes.

3D sketching in immersive environments: Shape from disordered ribbon strokes

Immersive environments with head mounted displays (HMD) and hand-held controllers, either in Virtual or Augmented Reality (VR/AR), offer new possibilities for the creation of artistic 3D content. Some of them are exploited by mid-air drawing applications: the user’s hand trajectory generates a set of stylized curves or ribbons in space, giving the impression of painting or drawing in 3D. We propose a method to extend this approach to the sketching of surfaces with a VR controller. The idea is to favor shape exploration, offering a tool, where the user creates a surface just by painting ribbons. These ribbons are not constrained to form patch boundaries for example or to completely cover the shape. They can be very sparse, disordered, overlap or not, intersect or not. The shape is computed simultaneously, starting with the first piece of ribbon drawn by the user and continuing to evolve in real-time as long as the user continues sketching. Our method involves minimizing an energy function based on the projections of the ribbon strokes on a proxy surface by taking the controller’s orientations into account. The current implementation considers elevation surfaces. In addition to many examples, we evaluate the time performance of the dynamic shape modeling with respect to an increasing number of input ribbon strokes. Finally, we present images of an artistic creation that combines stylized curve drawings in VR with our surface sketching tool created by a professional artist.

Semi-automated creation of reciprocal frame structures using deep learning

Systems that can transform two-dimensional (2D) sketches into 3D models while performing structural analyses are necessary for architectural sketches. To address this challenge, this paper focuses on how deep-learning algorithms can aid in this transformation process. It presents a model that uses the instance-segmentation technique with Mask RCNN to detect and distinguish two types of short beams of reciprocal frame structures (RFs) in 2D sketches and uses this information in the systematic creation of a 3D model of RFs to conduct their structural analysis. The results indicate that the model is capable of clustering beam types in 2D sketches via masking and classifying, eliminating irrelevant background objects, creating parametric RFs using masking information, and performing structural analysis. The model, which helps optimise and ease the design process, can be used by architects or engineers. This paper will inspire future work on the creation of integrated modelling systems.

Mobility of geometric constraint systems with extrusion symmetry

AbstractIf we take a (bar-joint) framework, prepare an identical copy of this framework, translate it by some vector $$\tau $$ τ , and finally join corresponding points of the two copies, then we obtain a framework with ‘extrusion’ symmetry in the direction of $$\tau $$ τ . This process may be repeated t times to obtain a framework whose underlying graph has $$\mathbb {Z}_2^t$$ Z 2 t as a subgroup of its automorphism group and which has ‘t-fold extrusion’ symmetry. Extruding a framework is a widely used technique in CAD for generating a 3D model from an initial 2D sketch, and hence it is important to understand the flexibility of extrusion-symmetric frameworks. Using group representation theory, we show that while t-fold extrusion symmetry is not a point-group symmetry, the rigidity matrix of a framework with t-fold extrusion symmetry can still be transformed into a block-decomposed form in the analogous way as for point-group symmetric frameworks. This allows us to establish Fowler-Guest-type character counts to analyse the mobility of such frameworks. We show that this entire theory also extends to the more general point-hyperplane frameworks with t-fold extrusion symmetry. Moreover, we show that under suitable regularity conditions the infinitesimal flexes we detect with our symmetry-adapted counts extend to finite (continuous) motions. Finally, we establish an algorithm that checks for finite motions via linearly displacing framework points along velocity vectors of infinitesimal motions.

Fostering computational thinking with simulated 3D robots in secondary education

AbstractComputational Thinking (CT) can be defined as the thought processes involved in formulating problems so that their solutions can be represented as sequential steps and algorithms. It is a key skill for children in the 21st century. However, it is unclear how CT can be developed most effectively in children. Several pedagogical methodologies have been proposed and are being investigated. The main aim of this paper is to test the hypothesis which states that using three‐dimensional (3D) simulated robots helps in the learning of programming and CT concepts, such as directions, loops, conditionals, and functions. The research questions are: Does this hypothesis hold true? Are some concepts easier or better learned than others and to what extent? The goal is to measure and evaluate the effect of using as a learning tool a platform with 3D simulated robots and realistic physics, and compare it with the standard Scratch learning tool which does not use robotics but a two‐dimensional (2D) cartoon avatar they are already familiar with. For practical reasons, a quasiexperimental design with nonequivalent groups and 85 second‐year Secondary Education students (ages 12–13) was performed. They were separated into control and experimental groups and followed a seven‐session intervention with the baseline 2D Scratch and the 3D simulated robots platform, respectively. Both used a visual block programming language and the same activities. To have quantitative and reliable results, a widely accepted CT test has been used, pre‐ and postintervention. Also qualitative feedback is presented. The obtained results show that using the platform with simulated 3D robots significantly helps when developing students' CT. With it, the students do learn basic programming concepts and reach higher scores in the CT test. This improvement applies to all CT‐analyzed concepts except in functions where the grades are maintained. Furthermore, students manage to master the activities on the 3D simulated robots platform, which reflects on the empowerment the platform has got in them.

Efficacy of a Virtual 3D Simulation-Based Digital Training Module for Building Dental Technology Students' Long-Term Competency in Removable Partial Denture Design: Prospective Cohort Study.

Removable partial denture (RPD) design is crucial to long-term success in dental treatment, but shortcomings in RPD design training and competency acquisition among dental students have persisted for decades. Digital production is increasing in prevalence in stomatology, and a digital RPD (D-RPD) module, under the framework of the certified Objective Manipulative Skill Examination of Dental Technicians (OMEDT) system reported in our previous work, may improve on existing RPD training models for students. We aimed to determine the efficacy of a virtual 3D simulation-based progressive digital training module for RPD design compared to traditional training. We developed a prospective cohort study including dental technology students at the Stomatology College of Chongqing Medical University. Cohort 1 received traditional RPD design training (7 wk). Cohort 2 received D-RPD module training based on text and 2D sketches (7 wk). Cohort 3 received D-RPD module pilot training based on text and 2D sketches (4 wk) and continued to receive training based on 3D virtual casts of real patients (3 wk). RPD design tests based on virtual casts were conducted at 1 month and 1 year after training. We collected RPD design scores and the time spent to perform each assessment. We collected the RPD design scores and the time spent to perform each assessment at 1 month and 1 year after training. The study recruited 109 students, including 58 (53.2%) female and 51 male (56.8%) students. Cohort 1 scored the lowest and cohort 3 scored the highest in both tests (cohorts 1-3 at 1 mo: mean score 65.8, SD 21.5; mean score 81.9, SD 6.88; and mean score 85.3, SD 8.55, respectively; P<.001; cohorts 1-3 at 1 y: mean score 60.3, SD 16.7; mean score 75.5, SD 3.90; and mean score 90.9, SD 4.3, respectively; P<.001). The difference between cohorts in the time spent was not statistically significant at 1 month (cohorts 1-3: mean 2407.8, SD 1370.3 s; mean 1835.0, SD 1329.2 s; and mean 1790.3, SD 1195.5 s, respectively; P=.06) but was statistically significant at 1 year (cohorts 1-3: mean 2049.16, SD 1099.0 s; mean 1857.33, SD 587.39 s; and mean 2524.3, SD 566.37 s, respectively; P<.001). Intracohort comparisons indicated that the differences in scores at 1 month and 1 year were not statistically significant for cohort 1 (95% CI -2.1 to 13.0; P=.16), while cohort 3 obtained significantly higher scores 1 year later (95% CI 2.5-8.7; P=.001), and cohort 2 obtained significantly lower scores 1 year later (95% CI -8.8 to -3.9; P<.001). Cohort 3 obtained the highest score at both time points with retention of competency at 1 year, indicating that progressive D-RPD training including virtual 3D simulation facilitated improved competency in RPD design. The adoption of D-RPD training may benefit learning outcomes.

Multi3D: 3D-aware multimodal image synthesis

3D-aware image synthesis has attained high quality and robust 3D consistency. Existing 3D controllable generative models are designed to synthesize 3D-aware images through a single modality, such as 2D segmentation or sketches, but lack the ability to finely control generated content, such as texture and age. In pursuit of enhancing user-guided controllability, we propose Multi3D, a 3D-aware controllable image synthesis model that supports multi-modal input. Our model can govern the geometry of the generated image using a 2D label map, such as a segmentation or sketch map, while concurrently regulating the appearance of the generated image through a textual description. To demonstrate the effectiveness of our method, we have conducted experiments on multiple datasets, including CelebAMask-HQ, AFHQ-cat, and shapenet-car. Qualitative and quantitative evaluations show that our method outperforms existing state-of-the-art methods.

Synthesis, Style Editing, and Animation of 3D Cartoon Face

Synthesis, Style Editing, and Animation of 3D Cartoon Face

How three-dimensional sketching environments affect spatial thinking: A functional magnetic resonance imaging study of virtual reality.

Designers rely on sketching to visualize and refine their initial ideas, and virtual reality (VR) tools now facilitate sketching in immersive 3D environments. However, little research has been conducted on the differences in the visual and spatial processes involved in 3D versus 2D sketching and their effects on cognition. This study investigated potential differences in spatial and visual functions related to the use of 3D versus 2D sketching media by analyzing functional magnetic resonance imaging (fMRI) data. We recruited 20 healthy, right-handed students from the Department of Horticulture and Landscape Architecture with at least three years of experience in freehand landscape drawing. Using an Oculus Quest VR headset controller and a 12.9-inch iPad Pro with an Apple Pencil, we tested participants individually with 3D and 2D sketching, respectively. When comparing 2D and 3D sketches, our fMRI results revealed significant differences in the activation of several brain regions, including the right middle temporal gyrus, both sides of the parietal lobe, and the left middle occipital gyrus. We also compared different sketching conditions, such as lines, geometrical objects (cube), and naturalistic objects (perspective view of a tree), and found significant differences in the spatial and visual recognition of brain areas that support visual recognition, composition, and spatial perception. This finding suggests that 3D sketching environments, such as VR, may activate more visual-spatial functions during sketching compared to 2D environments. The result highlights the potential of immersive sketching environments for design-related processes and spatial thinking.

HABERTAN: Game Petualangan 3D Dengan Tema Pemilahan Sampah Sebagai Upaya Pendekatan Inovatif Untuk Pengenalan Lingkungan

The increasing awareness of the importance of maintaining forest cleanliness as a primary ecosystem is becoming more urgent, especially in remote areas with limited deep socialization. This research aims to develop an innovative game titled "HABERTAN" focusing on waste sorting in forest areas to raise public awareness of the waste present in the forest. By utilizing a 3D model for visual assets with stages of 2D sketch, 3D modeling, UV Mapping, and Texture Baking. The development of this game is conducted using the Analysis, Design, Development, Implementation, and Evaluation (ADDIE) method. The ADDIE method is used to ensure a systematic game development process, starting from needs analysis to performance evaluation. The goal of this game is to provide players with an interactive experience that not only entertains but also provides a deep understanding of the impact of waste on forest sustainability. The game "HABERTAN (Let's Clean the Forest)" is designed to provide players with a fun and educational learning experience. The results of the development of this game show that "HABERTAN" provides a unique experience for players, with the ability to learn while playing. It is hoped that this game can be an effective tool in raising public awareness of the importance of maintaining forest cleanliness, especially in the context of waste sorting. Through this innovative approach, it is hoped to encourage active participation from the community in preserving forests as extremely valuable natural resources.

Visual thinking in virtual environments: evaluating multidisciplinary interaction through drawing ideation in real-time remote co-design

This study investigated remote multidisciplinary sketching ideation across three systems: virtual reality (VR), tablet drawing, and uploading images of paper drawings. Though cumbersome, expressiveness and line control with drawing on paper was still noted to be important even in remote sketching, particularly by people experienced with this method. The tablet method was user-friendly, fostering effective collaborative understanding, especially in object-based ideation. Existing skills played a significant role in shaping collaborative perceptions. Despite challenges, VR exhibited promise in fostering creative expression and visualization in collaborative design workflows. Notably, it proved beneficial in early problem-solving stages where spatial and sensory considerations influenced structural decisions—potentially useful after general brainstorming and 2D sketching has established themes and objects. This research contributes to further understanding of VR’s evolving role in design thinking, its synergy with other drawing methods in remote sketching collaboration, and the evolving landscape of diverse user needs in ideation processes.