Research map / Image generation / Text-to-image diffusion
Towards Training-free Scene Text Editing
Yubo Li, Xugong Qin, Peng Zhang, Hailun Lin, Gangyan Zeng, Kexin Zhang
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 FMS module to the editing pipeline (ablation) | TextFlow (FLUX-Kontext) · ScenePair | PSNR: 20.47 -> 22.42 | Improved |
| add FMS module to the editing pipeline (ablation) | TextFlow (FLUX-Kontext) · ScenePair | FID: 16.69 -> 13.52 | Improved |
| add AttnBoost to the editing pipeline (ablation) | TextFlow (FLUX-Kontext) · ScenePair | ACC (%): 20.35 -> 79.80 | Improved |
| add AttnBoost to the editing pipeline (ablation) | TextFlow (FLUX-Kontext) · ScenePair (Random) | ACC (%): 18.84 -> 74.52 | Improved |
| TextFlow vs plain FLUX-Kontext baseline | FLUX-Kontext · ScenePair (Random) | ACC (%): 76.63 -> 74.52 | Hurt |
| use Overshoot scheduler instead of Euler (ablation) | TextFlow · ScenePair | ACC (%): 78.73 -> 79.90 | Improved |
| increase denoising steps from 50 to 70 (ablation) | TextFlow · ScenePair | FID: 13.53 -> 12.83 | No clear effect |
| increase denoising steps from 24 to 50 (ablation) | TextFlow · ScenePair | ACC (%): 77.97 -> 79.98 | Improved |
| use TextFlow instead of training-based TextFlux | TextFlow vs TextFlux · ScenePair | ACC (%): 80.40 -> 79.98 ACC; FID 54.64 -> 13.53 | Mixed |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- FlowEdit — FlowEdit: Inversion-Free Text-Based Editing Using Pre-Trained Flow Models (ICCV 2025) · better on 4 of 4 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
- SteerVTE: Seamless Video Text Editing with Style and Glyph Control (NeurIPS 2026 · similarity 0.89)
- TextCtrl: Diffusion-based Scene Text Editing with Prior Guidance Control (NeurIPS 2024 · similarity 0.89)
- Text-Driven Image Editing via Learnable Regions (CVPR 2024 · similarity 0.87)
- Text-Vision Co-Instructed Image Editing (NeurIPS 2026 · similarity 0.87)
- DreamText: High Fidelity Scene Text Synthesis (CVPR 2025 · similarity 0.87)
- AdapEdit: Spatio-Temporal Guided Adaptive Editing Algorithm for Text-Based Continuity-Sensitive Image Editing (AAAI 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.
