Research map / Image generation / Autoregressive and latent generation
Picture That Sketch: Photorealistic Image Generation From Abstract Sketches
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 |
|---|---|---|---|
| Replace autoregressive mapping with the baseline latent mapper (ablation) | Picture That Sketch · ShoeV2 | FID: 35.85 -> 57.27 | Hurt |
| Replace autoregressive mapping with the baseline latent mapper (ablation) | Picture That Sketch · ChairV2 | FID: 90.21 -> 111.99 | Hurt |
| Remove fine-grained FG-SBIR discriminative loss (ablation) | Picture That Sketch · ShoeV2 | FID: 35.85 -> 50.29 | Hurt |
| Remove fine-grained FG-SBIR discriminative loss (ablation) | Picture That Sketch · ChairV2 | FID: 90.21 -> 104.29 | Hurt |
| Remove the photo-to-photo teacher E_r (ablation) | Picture That Sketch · ShoeV2 | FID: 35.85 -> 46.87 | Hurt |
| Remove the photo-to-photo teacher E_r (ablation) | Picture That Sketch · ChairV2 | FID: 90.21 -> 99.38 | Hurt |
| Use naive W+ encoding instead of the proposed mapper (ablation) | Picture That Sketch · ShoeV2 | FID: 35.85 -> 46.87 | Hurt |
| Use naive W+ encoding instead of the proposed mapper (ablation) | Picture That Sketch · ChairV2 | FID: 90.21 -> 99.24 | Hurt |
| Embed the sketch as a single W latent vector (ablation) | Picture That Sketch · ShoeV2 | FID: 35.85 -> 52.35 | Hurt |
| Embed the sketch as a single W latent vector (ablation) | Picture That Sketch · ChairV2 | FID: 90.21 -> 107.99 | Hurt |
Closest papers by meaning
- SketchingReality: From Freehand Scene Sketches to Photorealistic Images (ICLR 2026 · similarity 0.92)
- Sketch Down the FLOPs: Towards Efficient Networks for Human Sketch (CVPR 2025 · similarity 0.91)
- It's All About Your Sketch: Democratising Sketch Control in Diffusion Models (CVPR 2024 · similarity 0.91)
- How to Handle Sketch-Abstraction in Sketch-Based Image Retrieval? (CVPR 2024 · similarity 0.90)
- SketchFusion: Learning Universal Sketch Features through Fusing Foundation Models (CVPR 2025 · similarity 0.88)
- One-Shot Reference-based Structure-Aware Image to Sketch Synthesis (AAAI 2025 · similarity 0.88)
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.