cs.CVOct 30, 2026

SketchDirector: Complex Scene Sketching via Agentic Layout Reasoning and Detail Refinement

Authors: Yiqi Liang, Xiaodong Cun

Organizations: GVC Lab, Great Bay University

Abstract

Existing sketch generation methods often rely on image guidance and are largely limited to individual objects or simplified scenes, leaving text-guided complex scene sketching underexplored. We present SketchDirector, an agentic framework that mimics the human sketching workflow for automatic multi-concept scene sketching. Given a scene-level text prompt, SketchDirector first employs a large language model~(LLM) to decompose the scene into multiple draft assets, where each is represented by a set of Bézier curves. Then, an agentic reasoning loop iteratively plans and executes their spatial arrangement, guided by self-evolving memory. Finally, a detail refinement module leverages diffusion priors to enhance sketch quality while preserving geometric structure. To support future research and applications in multi-concept scene sketch generation, we further introduce the MCSketch8K dataset, comprising editable vector assets, compositional layouts, and object-level manipulation annotations. Extensive experiments show that SketchDirector outperforms state-of-the-art baselines in semantic and compositional quality while effectively preserving overall contour organization during refinement.

Figures & tables

Explore similar work

CardsList
  1. 3DrawAgent: Teaching LLM to Draw in 3D with Early Contrastive Experience

    Apr 9, 2026Hongcan Xiao, Xinyue Xiao, Yilin Wang +23D Spatial ReasoningRL for Generative Models

  2. Geometrically Consistent Multi-View Scene Generation from Freehand Sketches

    Apr 15, 2026Ahmed Bourouis, Savas Ozkan, Andrea Maracani +23D Scene GenerationMulti-View Consistency

  3. Teaching an Agent to Sketch One Part at a Time

    Mar 19, 2026Xiaodan Du, Ruize Xu, David Yunis +2Generative ModelingLLM Agent Training