Computer-Aided Design (CAD) generative modeling is driving significant innovations across industrial applications. Recent works have shown remarkable progress in creating solid models from various inputs such as point clouds, meshes, and text descriptions. However, these methods fundamentally diverge from traditional industrial workflows that begin with 2D engineering drawings. The automatic generation of parametric CAD models from these 2D vector drawings remains underexplored despite being a critical step in engineering design. To address this gap, our key insight is to reframe CAD generation as a sequence-to-sequence learning problem where vector drawing primitives directly inform the generation of parametric CAD operations, preserving geometric precision and design intent throughout the transformation process. We propose Drawing2CAD, a framework with three key technical components: a network-friendly vector primitive representation that preserves precise geometric information, a dual-decoder transformer architecture that decouples command type and parameter generation while maintaining precise correspondence, and a soft target distribution loss function accommodating inherent flexibility in CAD parameters. To train and evaluate Drawing2CAD, we create CAD-VGDrawing, a dataset of paired engineering drawings and parametric CAD models, and conduct thorough experiments to demonstrate the effectiveness of our method. Code and dataset are available at https://github.com/lllssc/Drawing2CAD.
Abstract Breast cancer is one of the most common malignancies among women globally. Magnetic resonance imaging (MRI), as the final non-invasive diagnostic tool before biopsy, provides detailed free-text reports that support clinical decision-making. Therefore, the effective utilization of the information in MRI reports to make reliable decisions is crucial for patient care. This study proposes a novel method for BI-RADS classification using breast MRI reports. Large language models are employed to transform free-text reports into structured reports. Specifically, missing category information (MCI) that is absent in the free-text reports is supplemented by assigning default values to the missing categories in the structured reports. To ensure data privacy, a locally deployed Qwen-Chat model is employed. Furthermore, to enhance the domain-specific adaptability, a knowledge-driven prompt is designed. The Qwen-7B-Chat model is fine-tuned specifically for structuring breast MRI reports. To prevent information loss and enable comprehensive learning of all report details, a fusion strategy is introduced, combining free-text and structured reports to train the classification model. Experimental results show that the proposed BI-RADS classification method outperforms existing report classification methods across multiple evaluation metrics. Furthermore, an external test set from a different hospital is used to validate the robustness of the proposed approach. The proposed structured method surpasses GPT-4o in terms of performance. Ablation experiments confirm that the knowledge-driven prompt, MCI, and the fusion strategy are crucial to the model’s performance.
Drawing. Design. Illustration, Computer applications to medicine. Medical informatics
This study examines how graphic design functions as a social infrastructure within high-rise urban housing, shaping interactions, identity, and collective well-being. Urbanization has driven vertical expansion, creating new spatial and social challenges that affect community cohesion. While existing research in architecture and urban sociology has focused on spatial design and technological efficiency, this study highlights the overlooked role of visual communication as an active mediator of social relations. The research introduces the concept of Vertical Social Infrastructures, which reframes high-rise housing as a visual–social system rather than merely a physical structure. Using comparative analysis, the study explores how graphic elements, color, signage, murals, and typography can guide interaction and create shared narratives in dense residential settings. The study explicitly contributes to the field of graphic design by demonstrating how visual communication — including color, signage, murals, and typography — mediates social interaction and spatial behavior in vertical housing environments. This connection reinforces the journal’s focus on the intersections between design, media, and society, positioning graphic design as both an analytical and infrastructural framework for community engagement. The findings demonstrate that graphic design serves as a connective infrastructure, enhancing spatial legibility and fostering social engagement. By integrating theories from design studies, social infrastructure, and urban communication, this paper contributes a cross-disciplinary framework for understanding design as a medium of social sustainability. The results suggest that visual design strategies can transform vertical housing into inclusive, participatory, and emotionally resonant environments.
We present the design, simulation and optimization of a graphene-on-silicon nitride (GOSiN) integrated waveguide dual-mode electro-absorption modulator operating with a speed between 27-109 GHz and an energy consumption below 6 pJ/bit. This device individually modulates the TE$_0$ and TE$_1$ modes in a single-arm dual-mode waveguide with modulation depths up to 316 dB/cm and 273 dB/cm, respectively. It has promising applications in Multimode Division Multiplexing (MDM) systems, where single-mode modulation induces high losses and costs. We have started from the design of GOSiN TE$_0$ and TE$_1$ passive low-loss mode filters. Then, applying a gate voltage to graphene via transparent IHO electrodes, we have combined both filters and shown that the light absorption can be modulated to obtain four logical values in transmission: (0,0), (0,1), (1,0) and (1,1). Our proposed devices can potentially boost the development of efficient MDM systems for ultra-fast on-chip interconnections.
This dossier is about comics as sound objects. The collection of articles and interviews examines how comics can be adapted, remediated and/or transformed into aural works. The collection’s scope includes songs, radio dramas, audiobooks, and other aural forms, such as a spoken reading of a comic.
Drawing. Design. Illustration, Literature (General)
Walter José Castañeda Marulanda, Silvana Lopez Bernal, William Ospina Toro
Este artículo presenta la pervivencia de procedimientos preindustriales en la obtención de artefactos, caso de estudio municipio de Manzanares, Caldas. La investigación se basa en los hallazgos del estudio titulado “Incidencia del diseño en el contexto regional: objetos, mensajes y ambientes anteriores a 1950”, realizado en 2007 en municipios del norte de Caldas. Para este estudio se utilizaron las metodologías de entrevista semiestructurada, muestreo aleatorio simple y categorización para el análisis de información. Como resultado se evidenció una ausencia de procesos industriales y de modelos metodológicos característicos de la actividad proyectual más institucionalizada, de allí que se manifestaron procesos artesanales que ofrecen soluciones de carácter utilitario, respondiendo principalmente a necesidades funcionales relacionadas con el bienestar particular. Estos hallazgos, que evidencian la pervivencia de procedimientos vernáculos de fabricación y maneras de hacer tradicionales, deberían ser considerados como parte de la historia de diseño en Colombia y como antecedentes importantes a tener en cuenta acerca de la adaptación del diseño como proceso y disciplina, los cuales aún persisten gracias a las manifestaciones objetuales de artesanos y artífices en algunas regiones rurales del país.
Dos Punts, un estudio de diseño gráfico y comunicación fundado por Neus Seguí y Anna Brullas. El color, la ilustración y una comunicación propia, directa y cercana son el sello de identidad en todos nuestros proyectos. Trabajamos de la mano con el cliente y colaboradores externos para llegar a soluciones gráficas a la medida de lasnecesidades de cada encargo.
A lately created metaheuristic algorithm called Child Drawing Development Optimization (CDDO) has proven to be effective in a number of benchmark tests. A Binary Child Drawing Development Optimization (BCDDO) is suggested for choosing the wrapper features in this study. To achieve the best classification accuracy, a subset of crucial features is selected using the suggested BCDDO. The proposed feature selection technique's efficiency and effectiveness are assessed using the Harris Hawk, Grey Wolf, Salp, and Whale optimization algorithms. The suggested approach has significantly outperformed the previously discussed techniques in the area of feature selection to increase classification accuracy. Moderate COVID, breast cancer, and big COVID are the three datasets utilized in this study. The classification accuracy for each of the three datasets was (98.75, 98.83%, and 99.36) accordingly.
Sejak pandemi penggunaan teknologi dalam setiap aspek kehidupan meningkat. Dengan bantuan teknologi semua dapat diperoleh dengan cepat dan instan. Belanja dilakukan secara remote, termasuk belanja fashion. Retail bergeser dari bangunan fisik di mal menjadi antarmuka di genggaman handphone konsumen. Konsumen tidak perlu datang dan mencoba pakaian yang dibeli. Di Tokopedia atau Shopee, platform belanja online terbesar di Indonesia, segera setelah konsumen memilih pakaian mana yang disuka berdasarkan foto produk yang ditampilkan, konsumen dapat langsung membelinya. Lebih jauh lagi ada aplikasi Virtual Try-On (VTO) seperti Browzwear dan Lalaland yang memungkinkan konsumen mencoba langsung baju yang akan dibeli. Penelitian ini memberikan gambaran kepada pelaku bisnis fashion, akademisi, dan peneliti tentang bagaimana fashion retail mengkomunikasikan produknya kepada konsumennya dengan memanfaatkan Augmented Reality (AR)/Machine Learning (ML)/Computer Vision (CV) di era big data menggunakan dataset yang terdiri dari ribuan atau jutaan foto. Hal ini membuat proses produksi dan konsumsinya lebih cepat dan lebih hemat, sehingga implementasi AI juga dapat dipandang sebagai salah satu alternatif yang berkelanjutan. Metode penelitian terdiri dari dua tahap. Pertama yaitu dengan analisis jurnal-jurnal sains komputer serta investigasi aplikasi-aplikasi AR seperti Zero 10 dan software 3D seperti CLO atau Browzwear, dikaitkan dengan isu berkelanjutan dengan tujuan untuk mengidentifikasi cara kerja Virtual Try-On. Tahap selanjutnya dilakukan analisis terhadap cara kerja VTO dari perspektif komunikasi brand terhadap konsumen, agar dapat mendeskripsikan seperti apa retail berkelanjutan di era big data.
We consider drawings of graphs in the plane in which vertices are assigned distinct points in the plane and edges are drawn as simple curves connecting the vertices and such that the edges intersect only at their common endpoints. There is an intuitive quality measure for drawings of a graph that measures the height of a drawing $φ: G \rightarrow \mathbb{R}^2$ as follows. For a vertical line $\ell$ in $\mathbb{R}^2$, let the height of $\ell$ be the cardinality of the set $\ell \cap φ(G)$. The height of a drawing of $G$ is the maximum height over all vertical lines. In this paper, instead of abstract graphs, we fix a drawing and consider plane graphs. In other words, we are looking for a homeomorphism of the plane that minimizes the height of the resulting drawing. This problem is equivalent to the homotopy height problem in the plane, and the homotopic Fréchet distance problem. These problems were recently shown to lie in NP, but no polynomial-time algorithm or NP-hardness proof has been found since their formulation in 2009. We present the first polynomial-time algorithm for drawing trees with optimal height. This corresponds to a polynomial-time algorithm for the homotopy height where the triangulation has only one vertex (that is, a set of loops incident to a single vertex), so that its dual is a tree.
Access to large and diverse computer-aided design (CAD) drawings is critical for developing symbol spotting algorithms. In this paper, we present FloorPlanCAD, a large-scale real-world CAD drawing dataset containing over 10,000 floor plans, ranging from residential to commercial buildings. CAD drawings in the dataset are all represented as vector graphics, which enable us to provide line-grained annotations of 30 object categories. Equipped by such annotations, we introduce the task of panoptic symbol spotting, which requires to spot not only instances of countable things, but also the semantic of uncountable stuff. Aiming to solve this task, we propose a novel method by combining Graph Convolutional Networks (GCNs) with Convolutional Neural Networks (CNNs), which captures both non-Euclidean and Euclidean features and can be trained end-to-end. The proposed CNN-GCN method achieved state-of-the-art (SOTA) performance on the task of semantic symbol spotting, and help us build a baseline network for the panoptic symbol spotting task. Our contributions are three-fold: 1) to the best of our knowledge, the presented CAD drawing dataset is the first of its kind; 2) the panoptic symbol spotting task considers the spotting of both thing instances and stuff semantic as one recognition problem; and 3) we presented a baseline solution to the panoptic symbol spotting task based on a novel CNN-GCN method, which achieved SOTA performance on semantic symbol spotting. We believe that these contributions will boost research in related areas.
This paper offers a retrospective of what we learnt from organizing the workshop *Ethical Considerations in Creative applications of Computer Vision* at CVPR 2021 conference and, prior to that, a series of workshops on *Computer Vision for Fashion, Art and Design* at ECCV 2018, ICCV 2019, and CVPR 2020. We hope this reflection will bring artists and machine learning researchers into conversation around the ethical and social dimensions of creative applications of computer vision.
In the past decades, many graph drawing techniques have been proposed for generating aesthetically pleasing graph layouts. However, it remains a challenging task since different layout methods tend to highlight different characteristics of the graphs. Recently, studies on deep learning based graph drawing algorithm have emerged but they are often not generalizable to arbitrary graphs without re-training. In this paper, we propose a Convolutional Graph Neural Network based deep learning framework, DeepGD, which can draw arbitrary graphs once trained. It attempts to generate layouts by compromising among multiple pre-specified aesthetics considering a good graph layout usually complies with multiple aesthetics simultaneously. In order to balance the trade-off, we propose two adaptive training strategies which adjust the weight factor of each aesthetic dynamically during training. The quantitative and qualitative assessment of DeepGD demonstrates that it is capable of drawing arbitrary graphs effectively, while being flexible at accommodating different aesthetic criteria.
El progreso tecnológico y el acumulado material en la tecnosfera, definen nuevas fronteras que se conforman a través de los rangos de movilidad y accesibilidad a la tecnología. Esas zonas liminales actúan como filtros conformando distintas centralidades y subalternidades (o periferias), inclusiones y exclusiones, como formas nuevas de colonialidad. La materia de la tecnosfera, traicionada por la linealidad de su propio progreso y por el estrago que causa la naturaleza, es desechada transformándose en un innovador sentido de ruina. El arte la libera de su obsolescencia y la transforma en su material para visibilizar desde su abandono aquellas fronteras que genera el propio mundo tecnológico. El artista busca catalogar sus acumulaciones y serializarlas. La forma-trayecto se transforma en el tipo de operación artística privilegiada. El arte usa al objeto tecnológico (en ruinas) no ya por su efímera funcionalidad, sino por las propiedades ruinológicas que adquiere a partir de su desuso.
Drawing. Design. Illustration, Communication. Mass media
Aniruddha Singhal, Ayush Kumar, Shivam Thukral
et al.
This paper presents a robotic system (\textit{Chitrakar}) which autonomously converts any image of a human face to a recognizable non-self-intersecting loop (Jordan Curve) and draws it on any planar surface. The image is processed using Mask R-CNN for instance segmentation, Laplacian of Gaussian (LoG) for feature enhancement and intensity-based probabilistic stippling for the image to points conversion. These points are treated as a destination for a travelling salesman and are connected with an optimal path which is calculated heuristically by minimizing the total distance to be travelled. This path is converted to a Jordan Curve in feasible time by removing intersections using a combination of image processing, 2-opt, and Bresenham's Algorithm. The robotic system generates $n$ instances of each image for human aesthetic judgement, out of which the most appealing instance is selected for the final drawing. The drawing is executed carefully by the robot's arm using trapezoidal velocity profiles for jerk-free and fast motion. The drawing, with a decent resolution, can be completed in less than 30 minutes which is impossible to do by hand. This work demonstrates the use of robotics to augment humans in executing difficult craft-work instead of replacing them altogether.
The problem of extending partial geometric graph representations such as plane graphs has received considerable attention in recent years. In particular, given a graph $G$, a connected subgraph $H$ of $G$ and a drawing $\mathcal{H}$ of $H$, the extension problem asks whether $\mathcal{H}$ can be extended into a drawing of $G$ while maintaining some desired property of the drawing (e.g., planarity). In their breakthrough result, Angelini et al. [ACM TALG 2015] showed that the extension problem is polynomial-time solvable when the aim is to preserve planarity. Very recently we considered this problem for partial 1-planar drawings [ICALP 2020], which are drawings in the plane that allow each edge to have at most one crossing. The most important question identified and left open in that work is whether the problem can be solved in polynomial time when $H$ can be obtained from $G$ by deleting a bounded number of vertices and edges. In this work, we answer this question positively by providing a constructive polynomial-time decision algorithm.
Why is it that we can recognize object identity and 3D shape from line drawings, even though they do not exist in the natural world? This paper hypothesizes that the human visual system perceives line drawings as if they were approximately realistic images. Moreover, the techniques of line drawing are chosen to accurately convey shape to a human observer. Several implications and variants of this hypothesis are explored.
In a recent article (Auer et al, Algorithmica 2016) it was claimed that every outer-1-planar graph has a planar visibility representation of area $O(n\log n)$. In this paper, we show that this is wrong: There are outer-1-planar graphs that require $Ω(n^2)$ area in any planar drawing. Then wegive a construction (using crossings, but preserving a given outer-1-planar embedding) that results in an orthogonal box-drawing with O(n log n) area and at most two bends per edge.