Research map / Image generation / Editing and style transfer
Target-aware Image Editing via Cycle-consistent Constraints
Yanghao Wang, Zhen Wang, 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 cycle-consistent noise optimization (rounds 0 to 100) instead of random noise (ablation) | FlowCycle · PIE-Bench | source consistency | Improved |
| use target-aware optimized intermediate state instead of a transferred state (ablation) | FlowCycle · white dog edits (qualitative) | edit quality | Improved |
| use transferred target-aware state instead of random noise state (ablation) | FlowCycle · white dog edits (qualitative) | edit quality | Improved |
| use target-aware states for large editing deviations (ablation) | FlowCycle · white dog to lion with crown (qualitative) | background preservation | Improved |
| strong step-schedule acceleration (ablation) | FlowCycle · PIE-Bench | Structure Distance / time: time 247 -> 90 s; distance 0.013 -> 0.017; CLIP 25.48 -> 25.31 | Mixed |
| mid step-schedule acceleration (ablation) | FlowCycle · PIE-Bench | time / quality: time 247 -> 126 s; distance 0.013 -> 0.014; CLIP 25.48 -> 25.34 | Mixed |
| FlowCycle vs FlowAlign | FlowCycle · PIE-Bench | Structure Distance: 0.028 -> 0.013 (time 4 -> 247 s) | 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.
- FlowAlign — FlowAlign: Trajectory-Regularized, Inversion-Free Flow-based Image Editing (ICLR 2026) · better on 2 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
- SplitFlow: Flow Decomposition for Inversion-Free Text-to-Image Editing (NeurIPS 2025 · similarity 0.90)
- Preserve or Modify? Context-Aware Evaluation for Balancing Preservation and Modification in Text-Guided Image Editing (CVPR 2025 · similarity 0.88)
- Balancing Preservation and Modification: A Region and Semantic Aware Metric for Instruction-Based Image Editing (ICML 2025 · similarity 0.88)
- Invertible Consistency Distillation for Text-Guided Image Editing in Around 7 Steps (NeurIPS 2024 · similarity 0.87)
- Edit-R2: Context-Aware Reinforcement Learning for Multi-Turn Image Editing (NeurIPS 2026 · similarity 0.87)
- Improving Editability in Image Generation with Layer-wise Memory (CVPR 2025 · similarity 0.87)
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.
