Research map / Image generation / Text-to-image diffusion
SVGDreamer++: Advancing Editability and Diversity in Text-Guided SVG Generation
Ximing Xing, Qian Yu, Chuang Wang, Haitao Zhou, Jing Zhang, Dong Xu
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 Reward Feedback Learning (ReFL) to VPSD (ablation) | VPSD · 224x224, 128 paths | iteration steps / time: 500 -> 300 steps; 13m15s -> 6m45s | Improved |
| add ReFL to VPSD (ablation) | VPSD · 600x600, 256 paths | iteration steps / time: 500 -> 300 steps; 14m21s -> 7m21s | Improved |
| add ReFL to VPSD (ablation) | VPSD · text-to-SVG | aesthetic score | Improved |
| use VPSD instead of SDS (SVGDreamer scratch vs VectorFusion scratch) | SVG optimization · Text-to-SVG comparison | FID: 80.76 -> 39.30 | Improved |
| use VPSD instead of SDS (PSNR) | SVG optimization · Text-to-SVG comparison | PSNR: 6.33 -> 10.51 | Improved |
| add image vectorization step (SIVE) to SVGDreamer | SVGDreamer · Text-to-SVG comparison | FID: 39.30 -> 30.10 | Improved |
| upgrade SVGDreamer to SVGDreamer++ (HIVE and adaptive primitive control) | SVGDreamer · Text-to-SVG comparison | FID: 30.10 -> 22.13 | Improved |
| upgrade SVGDreamer to SVGDreamer++ | SVGDreamer · Text-to-SVG comparison | PSNR: 14.54 -> 15.80 | Improved |
| increase number of vector particles from 1 to 16 (ablation) | VPSD · astronaut riding a horse prompt | diversity | Improved |
| add HIVE instead of LIVE or SIVE (ablation) | SVGDreamer++ · qualitative vectorization | boundary accuracy and path control | Improved |
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.
- SVGDreamer — SVGDreamer: Text Guided SVG Generation with Diffusion Model (CVPR 2024) · better on 6 of 6 shared results
- DiffSketcher — DiffSketcher: Text Guided Vector Sketch Synthesis through Latent Diffusion Models (NeurIPS 2023) · better on 6 of 6 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
- SVGDreamer: Text Guided SVG Generation with Diffusion Model (CVPR 2024 · similarity 0.94)
- OmniSVG: A Unified Scalable Vector Graphics Generation Model (NeurIPS 2025 · similarity 0.90)
- Chat2SVG: Vector Graphics Generation with Large Language Models and Image Diffusion Models (CVPR 2025 · similarity 0.90)
- SVGBuilder: Component-Based Colored SVG Generation with Text-Guided Autoregressive Transformers (AAAI 2025 · similarity 0.89)
- StarVector: Generating Scalable Vector Graphics Code from Images and Text (AAAI 2025 · similarity 0.88)
- VFIG: Vectorizing Complex Figures in SVG with Vision-Language Models (NeurIPS 2026 · 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.
