Research map / Image generation / Editing and style transfer
FlowDC: Flow-Based Decoupling-Decay for Complex Image Editing
Yilei Jiang, Zhen Wang, Yanghao Wang, Jun Yu, Yueting Zhuang, Jun Xiao, Long Chen
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 Velocity Orthogonal Decay (ablation) | FlowDC · Complex-PIE-Bench | DINO: 47.58 -> 71.69 | Improved |
| add Velocity Orthogonal Decay (ablation) | FlowDC · Complex-PIE-Bench | LPIPS: 49.89 -> 23.72 | Improved |
| add Velocity Orthogonal Decay (ablation) | FlowDC · Complex-PIE-Bench | CLIP-T: 29.23 -> 27.69 | Hurt |
| add Progressive Semantic Orthogonalization (ablation) | FlowDC · Complex-PIE-Bench | CLIP-T: 27.30 -> 27.69 | Improved |
| add Progressive Semantic Orthogonalization (ablation) | FlowDC · Complex-PIE-Bench | LPIPS: 24.19 -> 23.72 | Improved |
| add Progressive Semantic Orthogonalization (ablation) | FlowDC · Complex-PIE-Bench | CLIP-I: 87.76 -> 87.72 | No clear effect |
| FlowDC vs FlowEdit | FlowDC · PIE-Bench++ | CLIP-T: 25.12 -> 25.13 | No clear effect |
| FlowDC vs RF-Edit | FlowDC · Complex-PIE-Bench | CLIP-T: 28.79 -> 27.69 | Hurt |
| FlowDC vs FlowEdit | FlowDC · Complex-PIE-Bench | DINO: 70.84 -> 71.69 | Improved |
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 8 of 8 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
- Target-aware Image Editing via Cycle-consistent Constraints (ECCV 2026 · similarity 0.90)
- FlowAlign: Trajectory-Regularized, Inversion-Free Flow-based Image Editing (ICLR 2026 · similarity 0.90)
- SplitFlow: Flow Decomposition for Inversion-Free Text-to-Image Editing (NeurIPS 2025 · similarity 0.90)
- h-Flow: Flexible Flow-based Image Editing via Doob's h-Transform (ECCV 2026 · similarity 0.89)
- Unveil Inversion and Invariance in Flow Transformer for Versatile Image Editing (CVPR 2025 · similarity 0.86)
- Safe Few-Step Generation via Velocity Editing (NeurIPS 2026 · 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.
