Research map / Image generation / Editing and style transfer
UIP2P: Unsupervised Instruction-based Image Editing via Edit Reversibility Constraint
Enis Simsar, Alessio Tonioni, Yongqin Xian, Thomas Hofmann, Federico Tombari
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 L_sim loss to base (ablation) | UIP2P · MagicBrush test | L1: 0.117 -> 0.089 | Improved |
| add L_sim loss to base (ablation) | UIP2P · MagicBrush test | DINO: 0.806 -> 0.872 | Improved |
| add L_attn loss on top of L_sim (ablation) | UIP2P · MagicBrush test | L1: 0.089 -> 0.062 | Improved |
| add L_attn loss on top of L_sim (ablation) | UIP2P · MagicBrush test | DINO: 0.872 -> 0.904 | Improved |
| add L_sim and L_attn to base (ablation) | UIP2P · MagicBrush test | CLIP-T: 0.309 -> 0.296 | Hurt |
| use random timestep for reverse edit instead of same t (ablation) | UIP2P · MagicBrush test | L1: 0.074 -> 0.062 | Improved |
| use random timestep for reverse edit instead of same t (ablation) | UIP2P · MagicBrush test | DINO: 0.889 -> 0.904 | Improved |
| use random timestep for reverse edit instead of same t (ablation) | UIP2P · MagicBrush test | CLIP-T: 0.299 -> 0.296 | Hurt |
| train on CC12M instead of IP2P dataset | UIP2P · MagicBrush test (single-turn) | L1: 0.0722 -> 0.0619 | Improved |
| train on CC12M instead of IP2P dataset | UIP2P · MagicBrush test (single-turn) | DINO: 0.8876 -> 0.9039 | Improved |
| use UIP2P instead of IP2P | UIP2P vs IP2P · PIE benchmark | Distance: 57.91 -> 27.05 | 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.
- InstructPix2Pix — InstructPix2Pix: Learning To Follow Image Editing Instructions (CVPR 2023) · better on 9 of 9 shared results
- Emu Edit — Emu Edit: Precise Image Editing via Recognition and Generation Tasks (CVPR 2024) · better on 2 of 2 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
- SuperEdit: Rectifying and Facilitating Supervision for Instruction-Based Image Editing (ICCV 2025 · similarity 0.90)
- InsightEdit: Towards Better Instruction Following for Image Editing (CVPR 2025 · similarity 0.89)
- AnyEdit: Mastering Unified High-Quality Image Editing for Any Idea (CVPR 2025 · similarity 0.89)
- Balancing Preservation and Modification: A Region and Semantic Aware Metric for Instruction-Based Image Editing (ICML 2025 · similarity 0.88)
- UltraEdit: Instruction-based Fine-Grained Image Editing at Scale (NeurIPS 2024 · similarity 0.87)
- PairEdit: Learning Semantic Variations for Exemplar-based Image Editing (NeurIPS 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.
