Research map / Image generation / Editing and style transfer
Self-Supervised Sketch-to-Image Synthesis
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 adversarial refinement after the AE | Ours AE · CelebA-HQ | FID: 25.9 -> 13.6 | Helped |
| add adversarial refinement after the AE | Ours AE · WikiArt | FID: 74.2 -> 32.6 | Helped |
| add content triplet loss (ablation) | Vanilla AE · CelebA-HQ | FID: 44.3 -> 34.8 | Helped |
| add style triplet loss (ablation) | Vanilla AE · CelebA-HQ | FID: 44.3 -> 35.7 | Helped |
| add style classification loss (ablation) | Vanilla AE · CelebA-HQ | FID: 44.3 -> 36.4 | Helped |
| add style and content classification losses (ablation) | Vanilla AE · CelebA-HQ | FID: 44.3 -> 34.7 | Helped |
| use all self-supervised objectives instead of reconstruction loss alone (ablation) | Vanilla AE · CelebA-HQ | FID: 44.3 -> 25.9 | Helped |
| add original DMI using raw sketch masks (ablation) | AE · WikiArt | LPIPS: 24.15 -> 20.7 | Helped |
| replace raw-sketch DMI masks with feature-map masks (ablation) | AE + DMI · WikiArt | LPIPS: 20.7 -> 18.6 | Helped |
| use channel-wise multiplication instead of full AdaIN (ablation) | AE · WikiArt | LPIPS: 22.3 -> 21.6 | Helped |
| use channel-wise multiplication instead of addition only (ablation) | AE · WikiArt | LPIPS: 23.1 -> 21.6 | Helped |
| train AE and GAN jointly rather than in two stages (ablation) | Ours AE+GAN · image synthesis | synthesis quality | Hurt |
Closest papers by meaning
- Picture That Sketch: Photorealistic Image Generation From Abstract Sketches (CVPR 2023 · similarity 0.90)
- One-Shot Reference-based Structure-Aware Image to Sketch Synthesis (AAAI 2025 · similarity 0.88)
- Sketch Down the FLOPs: Towards Efficient Networks for Human Sketch (CVPR 2025 · similarity 0.87)
- Text to Sketch Generation with Multi-Styles (NeurIPS 2025 · similarity 0.87)
- SketchINR: A First Look into Sketches as Implicit Neural Representations (CVPR 2024 · similarity 0.87)
- It's All About Your Sketch: Democratising Sketch Control in Diffusion Models (CVPR 2024 · similarity 0.87)
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.