Research map / Image generation / Editing and style transfer
Open-Edit: Open-Domain Image Manipulation with Open-Vocabulary Instructions
Xihui Liu, Zhe Lin, Jianming Zhang, Handong Zhao, Quan Tran, Xiaogang Wang, Hongsheng Li
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 edge constraints to the image decoder (ablation) | Open-Edit · User study (60 images, 10 users) | visual quality score: 1.3 -> 4.1 | Improved |
| add edge constraints to the image decoder (ablation) | Open-Edit · User study (60 images, 10 users) | detail preservation score: 1.2 -> 3.7 | Improved |
| add edge constraints to the image decoder (ablation) | Open-Edit · User study (60 images, 10 users) | instruction match score: 4.0 -> 4.5 | Improved |
| add sample-specific optimization (ablation) | Open-Edit · User study (60 images, 10 users) | visual quality score: 4.1 -> 4.4 | Improved |
| add sample-specific optimization (ablation) | Open-Edit · User study (60 images, 10 users) | detail preservation score: 3.7 -> 4.3 | Improved |
| add sample-specific optimization (ablation) | Open-Edit · User study (60 images, 10 users) | instruction match score: 4.5 -> 4.5 | No clear effect |
| add edge constraints to the image decoder (ablation) | Open-Edit · Oxford-102 test | L2 error: 0.19 -> 0.10 | Improved |
| add sample-specific optimization (ablation) | Open-Edit · Oxford-102 test | L2 error: 0.10 -> 0.05 | Improved |
| add edge constraints to the image decoder (ablation) | Open-Edit · Conceptual Captions val | LPIPS: 0.33 -> 0.17 | Improved |
| add sample-specific optimization (ablation) | Open-Edit · Conceptual Captions val | LPIPS: 0.17 -> 0.06 | Improved |
| add sample-specific optimization (ablation) | Open-Edit · Conceptual Captions val | L2 error: 0.12 -> 0.07 | Improved |
| add soft instruction grounding maps (ablation) | Open-Edit · qualitative apple example | visual inspection | Improved |
Closest papers by meaning
- SpaceEdit: Learning a Unified Editing Space for Open-Domain Image Color Editing (CVPR 2022 · similarity 0.89)
- Learning to Manipulate Artistic Images (AAAI 2024 · similarity 0.85)
- FireEdit: Fine-grained Instruction-based Image Editing via Region-aware Vision Language Model (CVPR 2025 · similarity 0.85)
- Text-Driven Image Editing via Learnable Regions (CVPR 2024 · similarity 0.85)
- Object-Aware Inversion and Reassembly for Image Editing (ICLR 2024 · similarity 0.85)
- OmniEdit: Building Image Editing Generalist Models Through Specialist Supervision (ICLR 2025 · similarity 0.84)
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.
