Single-Line Drawing Generation via Semantics-Driven Optimization
2026-06-01 • Graphics
GraphicsComputer Vision and Pattern Recognition
AI summaryⓘ
The authors created a new way to automatically draw single-line sketches from either text descriptions or pictures. Their method uses a special math curve (called a URBS curve) to make sure the drawing is one continuous line, like a real artist might draw. They also add controls to adjust the detail and style of the drawing. Compared to other AI art tools, their system makes drawings that look nicer and better capture the single-line style. Plus, because the drawings are vector-based, they can be easily used for things like embroidery or laser cutting.
line drawingvector graphicssingle continuous strokeuniform rational B-spline (URBS)score distillation samplingtext-to-image generationloss functionsvectorizationfabrication processesoptimization
Authors
Tanguy Magne, Alexandre Binninger, Ruben Wiersma, Olga Sorkine-Hornung
Abstract
Line drawings are a highly expressive art form that requires the artist to abstract and distill the essence of their subject. We present the first semantics-driven method for automatically generating single-line drawings in vector format, guided either by a text prompt describing the concept or an input image depicting it. Our approach leverages score distillation sampling to optimize the parameters of a uniform rational B-spline (URBS) curve, ensuring that the drawing consists of a single continuous stroke by design. This representation provides fine-grained control over the level of detail, while additional loss terms allow us to steer the final artistic style. We demonstrate that our method outperforms state-of-the-art text-to-image models and optimization pipelines for this task, producing results that are both more aesthetically pleasing and more faithful to the style of continuous line drawing artists. Furthermore, because our method generates a vectorized curve, it directly supports downstream fabrication processes such as embroidery, laser engraving and wire bending. Our code and results are available at https://github.com/tanguymagne/SLDgen.