Research map / Image generation / Autoregressive and latent generation
CoGS: Controllable Generation and Search from Sketch and Style
Cusuh Ham, Gemma Canet Tarrés, Tu Bui, James Hays, Zhe Lin, John Collomosse
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 |
|---|---|---|---|
| add class label input (ablation) | CoGS (sketch, style, codebook loss) · Pseudosketches val | FID: 58.202 -> 52.992 | Improved |
| add style loss (ablation) | CoGS (sketch, style, label) · Pseudosketches val | FID: 52.992 -> 50.630 | Improved |
| add style loss (ablation) | CoGS (sketch, style, no label) · Pseudosketches val | FID: 58.202 -> 58.327 | No clear effect |
| add style loss (ablation) | CoGS (sketch, style, label) · Pseudosketches val | Structure distance: 3.043 -> 2.387 | Improved |
| add class label (ablation) | CoGS (codebook+style loss) · Pseudosketches val | Style distance: 1.129 -> 1.100 | Improved |
| add style loss (ablation) | CoGS (no label) · Pseudosketches val | Structure distance: 3.063 -> 2.823 | Improved |
| evaluate on hand-drawn sketches instead of pseudosketches | CoGS · Sketchy DB vs Pseudosketches | FID: 50.630 -> 81.820 | Hurt |
| compare CoGS to SketchyGAN | CoGS vs SketchyGAN · Pseudosketches val | FID: 213.762 -> 50.630 | Improved |
| compare CoGS to SketchyGAN | CoGS vs SketchyGAN · Pseudosketches val | LPIPS: 0.541 -> 0.521 | Hurt |
Later papers that compared with this method
1 later accepted paper reports CoGS as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- MaskSketch: Unpaired Structure-Guided Masked Image Generation (CVPR 2023 · better on 3 of 3 shared results)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- SketchyGAN — SketchyGAN: Towards Diverse and Realistic Sketch to Image Synthesis (CVPR 2018) · better on 5 of 7 shared results
- CoCosNetv2 — CoCosNet v2: Full-Resolution Correspondence Learning for Image Translation (CVPR 2021) · better on 3 of 3 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- One-Shot Reference-based Structure-Aware Image to Sketch Synthesis (AAAI 2025 · similarity 0.87)
- Text to Sketch Generation with Multi-Styles (NeurIPS 2025 · similarity 0.87)
- Self-Supervised Sketch-to-Image Synthesis (AAAI 2021 · similarity 0.87)
- Picture That Sketch: Photorealistic Image Generation From Abstract Sketches (CVPR 2023 · similarity 0.87)
- It's All About Your Sketch: Democratising Sketch Control in Diffusion Models (CVPR 2024 · similarity 0.86)
- Zero-Painter: Training-Free Layout Control for Text-to-Image Synthesis (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.
