Single-Line Drawing Generation via Semantics-Driven Optimization

2026-06-01

AI Summary

This work proposes a semantic-driven method for automatically generating single-line vector drawings that emulate artists’ abstract representations of object essence. Guided by either text prompts or reference images, the approach optimizes parameters of uniform rational B-spline (URBS) curves under the strict constraint of a single continuous stroke, while incorporating a style-control loss term to enable tunable trade-offs between detail fidelity and artistic style. The entire optimization process operates in vector space via score distillation sampling, eliminating the need for rasterization. To the best of our knowledge, this is the first method capable of producing semantically controllable, stylistically faithful, and fabrication-friendly single-line vector illustrations. It outperforms existing text-to-image and vector-based optimization approaches in both aesthetic quality and style preservation, and is directly deployable in physical fabrication workflows such as embroidery and laser engraving.

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.

Problem

• single-line drawing
• vector graphics
• semantic-driven generation
• continuous stroke
• artistic abstraction

Innovation

• semantics-driven optimization
• single-line drawing
• vector graphics
• score distillation sampling
• URBS curve