Research map / Image generation / Autoregressive and latent generation
Generating Sketches in a Hierarchical Auto-Regressive Process for Flexible Sketch Drawing Manipulation at Stroke-Level
Sicong Zang, Shuhui Gao, Zhijun Fang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| generate sketches via hierarchical auto-regressive stroke prediction (Sketch-HARP) vs SketchEdit | stroke replacement · human scoring, 51 volunteers | human score: 1.45 -> 2.09 | Improved |
| Sketch-HARP vs SketchEdit on stroke erasion | stroke erasion · human scoring | human score: 1.04 -> 2.30 | Improved |
| Sketch-HARP vs best baseline on reconstruction FID (DS1) | sketch reconstruction · QuickDraw DS1 | FID: 10.80 -> 9.96 (vs SP-gra2seq) | Improved |
| Sketch-HARP vs DC-gra2seq on reconstruction | sketch reconstruction · QuickDraw DS1 and DS2 | CLIP-S / retrieval: FID 12.83 -> 9.96 (DS1); DC-gra2seq uses images and is stronger on some metrics | Mixed |
Closest papers by meaning
- BézierSketch: A generative model for scalable vector sketches (ECCV 2020 · similarity 0.89)
- SketchKnitter: Vectorized Sketch Generation with Diffusion Models (ICLR 2023 · similarity 0.89)
- CoSE: Compositional Stroke Embeddings (NeurIPS 2020 · similarity 0.89)
- SketchAgent: Language-Driven Sequential Sketch Generation (CVPR 2025 · similarity 0.86)
- Text to Sketch Generation with Multi-Styles (NeurIPS 2025 · similarity 0.85)
- AirSketch: Generative Motion to Sketch (NeurIPS 2024 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
