Research map / 3D vision / Face and shape
Free-Hand Sketch Synthesis with Deformable Stroke Models
Yi Li, Yi-Zhe Song, Timothy M. Hospedales, Shaogang Gong
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 |
|---|---|---|---|
| train on professional Disney portrait sketches instead of TU-Berlin non-expert sketches | face sketch synthesis · face category | resemblance to input photo | Improved |
| use 90s abstraction level (more short strokes) instead of 30s level | Disney portrait, Artist E · Image-sketch similarity | ordering accuracy: 90.00% -> 56.67% | Hurt |
| use 90s abstraction level instead of 30s level | Disney portrait, Artist A · Image-sketch similarity | ordering accuracy: 76.67% -> 73.33% | No clear effect |
| raise semantic-stroke parameter eta_sem to 8 and 11 and switch off continuity term for 90s sketches | perceptual grouping · Disney 90s level | parameter setting | No clear effect |
| limited training sketch variation (duck, teapot) | deformable stroke model · Image-sketch similarity | ordering accuracy: 63.33% duck, 66.67% teapot | Hurt |
| synthesized sketches versus real sketches on recognition | shark category · Human sketch recognition | recognition rate: 60% -> 40% | Hurt |
| synthesized sketches versus real sketches on recognition | horse, duck, bicycle, teapot, face · Human sketch recognition | recognition rate: 86.25% -> 100% horse, 78.75% -> 100% duck | Improved |
Closest papers by meaning
- BézierSketch: A generative model for scalable vector sketches (ECCV 2020 · similarity 0.90)
- CoSE: Compositional Stroke Embeddings (NeurIPS 2020 · similarity 0.89)
- Modelling complex vector drawings with stroke-clouds (ICLR 2024 · similarity 0.89)
- Stroke2Sketch: Harnessing Stroke Attributes for Training-Free Sketch Generation (ICCV 2025 · similarity 0.89)
- Learning Deep Sketch Abstraction (CVPR 2018 · similarity 0.88)
- SketchINR: A First Look into Sketches as Implicit Neural Representations (CVPR 2024 · similarity 0.86)
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.
