Research map / Image generation / Editing and style transfer
Learning To Generate Line Drawings That Convey Geometry and Semantics
Caroline Chan, Frédo Durand, Phillip Isola
Check 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 |
|---|---|---|---|
| add geometry loss (ablation) | line drawing generator · Contour Drawings | user preference: 92.2% | Improved |
| add geometry loss (ablation) | line drawing generator · Anime | user preference: 48.3% | No clear effect |
| add CLIP semantic loss (ablation) | line drawing generator · Contour Drawings and Anime | user preference: 98.9%; 84.9% | Improved |
| add appearance (cycle consistency) loss (ablation) | line drawing generator · Contour Drawings and Anime | user preference: 87.0%; 64.9% | Improved |
| add geometry loss (ablation) | line drawing generator · Contour Drawings | predicted depth map MSE: 0.0530 -> 0.0506 | Improved |
| add geometry loss (ablation) | line drawing generator · Anime | predicted depth map MSE: 0.0400 -> 0.0418 | Hurt |
| use CLIP features instead of DeepLabv3 features for semantic matching (ablation) | semantic loss · line drawing generation | qualitative drawing quality | Improved |
| use Inception v3 Mixed 6b features rather than penultimate-layer features | geometry feature extraction · Line Drawings nearest-neighbor classification | accuracy: 24.5% -> 33.5% | Improved |
Closest papers by meaning
- Unsupervised Sketch to Photo Synthesis (ECCV 2020 · similarity 0.90)
- Quality Metric Guided Portrait Line Drawing Generation From Unpaired Training Data (TPAMI 2022 · similarity 0.89)
- Neural Contours: Learning to Draw Lines From 3D Shapes (CVPR 2020 · similarity 0.88)
- Learning Geometry-Aware Representations by Sketching (CVPR 2023 · similarity 0.88)
- From 2D CAD Drawings to 3D Parametric Models: A Vision-Language Approach (AAAI 2025 · similarity 0.86)
- Open Vocabulary Semantic Scene Sketch Understanding (CVPR 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.
