Research map / Image generation / Editing and style transfer
LayerEdit: Disentangled Multi-Object Editing via Conflict-Aware Multi-Layer Learning
Fengyi Fu, Mengqi Huang, Lei Zhang, Zhendong Mao
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 time-dependent region removing (vs fixed r) (ablation) | LayerEdit · OIR/LoMOE ablation | CLIP-T: 0.2386 -> 0.2762 | Improved |
| add time-dependent region removing (vs linear r) (ablation) | LayerEdit · OIR/LoMOE ablation | CLIP-T: 0.2560 -> 0.2762 | Improved |
| add intra-layer textual guidance (ablation) | LayerEdit · OIR/LoMOE ablation | CLIP-T: 0.2489 -> 0.2762 | Improved |
| add transparency learning (vs mask-guided fusion) (ablation) | LayerEdit · OIR/LoMOE ablation | CLIP-T: 0.2415 -> 0.2762 | Improved |
| add transparency learning (vs mask-guided fusion) (ablation) | LayerEdit · OIR/LoMOE ablation | CLIP-I: 0.8352 -> 0.8420 | Improved |
| add 1st structure loss term in transparency learning (ablation) | LayerEdit · OIR/LoMOE ablation | CLIP-T: 0.2536 -> 0.2762 | Improved |
| add 2nd non-negativity loss term (ablation) | LayerEdit · OIR/LoMOE ablation | CLIP-T: 0.2508 -> 0.2762 | Improved |
| add 3rd normalization loss term (ablation) | LayerEdit · OIR/LoMOE ablation | CLIP-T: 0.2692 -> 0.2762 | Improved |
| LayerEdit vs best prior multi-object editors | SDXL backbone · OIR/LoMOE | FID: 68.38 -> 60.13 | Improved |
| switch backbone from SDXL to FLUX | LayerEdit · OIR/LoMOE | FID: 60.13 -> 57.91 | Improved |
| switch backbone from SDXL to FLUX | LayerEdit · OIR/LoMOE | CLIP-I: 0.8420 -> 0.8408 | No clear effect |
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.
- ParallelEdits — ParallelEdits: Efficient Multi-Aspect Text-Driven Image Editing with Attention Grouping (NeurIPS 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
- Improving Editability in Image Generation with Layer-wise Memory (CVPR 2025 · similarity 0.86)
- 3DitScene: Editing Any Scene via Language-guided Disentangled Gaussian Splatting (ICLR 2025 · similarity 0.86)
- LayerCraft: Enhancing Text-to-Image Generation with CoT Reasoning and Layered Object Integration (NeurIPS 2025 · similarity 0.85)
- PairEdit: Learning Semantic Variations for Exemplar-based Image Editing (NeurIPS 2025 · similarity 0.85)
- EditSplat: Multi-View Fusion and Attention-Guided Optimization for View-Consistent 3D Scene Editing with 3D Gaussian Splatting (CVPR 2025 · similarity 0.85)
- SplitFlow: Flow Decomposition for Inversion-Free Text-to-Image Editing (NeurIPS 2025 · similarity 0.85)
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.
