Research map / Image generation / Editing and style transfer
SIMON: SImultaneous Multi-Object Navigation
Yifeng Zhu, Siyuan Huang, Jun Bao, Jun Yu, Buyu Liu
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 multi-object energy guidance to baseline (ablation) | SIMON baseline · 3D-FUTURE multi-object navigation (ablation) | FID: 28.403 -> 23.212 | Improved |
| add multi-object energy guidance to baseline (ablation) | SIMON baseline · 3D-FUTURE multi-object navigation (ablation) | navigation LPIPS x1e-1: 5.229 -> 4.752 | Improved |
| add multi-object energy guidance to baseline (ablation) | SIMON baseline · 3D-FUTURE multi-object navigation (ablation) | background LPIPS x1e-1: 1.757 -> 1.017 | Improved |
| add content-aware attention on top of energy guidance (ablation) | SIMON +EG · 3D-FUTURE multi-object navigation (ablation) | navigation LPIPS x1e-1: 4.752 -> 4.525 | Improved |
| add content-aware attention on top of energy guidance (ablation) | SIMON +EG · 3D-FUTURE multi-object navigation (ablation) | FID: 23.212 -> 22.786 | Improved |
| add content-aware attention on top of energy guidance (ablation) | SIMON +EG · 3D-FUTURE multi-object navigation (ablation) | background MSE x1e3: 0.235 -> 0.246 | Hurt |
| add latent initialization to +EG+CA (ablation) | SIMON +EG+CA · 3D-FUTURE multi-object navigation (ablation) | background MSE x1e3: 0.246 -> 0.209 | Improved |
| add latent initialization to +EG+CA (ablation) | SIMON +EG+CA · 3D-FUTURE multi-object navigation (ablation) | FID: 22.786 -> 22.826 | Hurt |
| add latent initialization (ablation) | SIMON · 3D-FUTURE single-object inpainting | reconstruction fidelity MSE x1e3: 3.155 -> 1.964 | Improved |
| add latent initialization (ablation) | SIMON · 3D-FUTURE single-object inpainting | residual suppression LPIPS: 0.996 -> 1.435 | Improved |
| add latent initialization (ablation) | SIMON · 3D-FUTURE single-object inpainting | reconstruction fidelity LPIPS: 0.991 -> 0.670 | Improved |
| simultaneous editing vs sequential single-object methods | SIMON vs DiffEditor · 3D-FUTURE multi-object navigation (1000 images) | FID: 30.193 -> 22.826 | 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.
- DiffEditor — DiffEditor: Boosting Accuracy and Flexibility on Diffusion-based Image Editing (CVPR 2024) · better on 5 of 5 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
- EditSplat: Multi-View Fusion and Attention-Guided Optimization for View-Consistent 3D Scene Editing with 3D Gaussian Splatting (CVPR 2025 · similarity 0.87)
- MOVIS: Enhancing Multi-Object Novel View Synthesis for Indoor Scenes (CVPR 2025 · similarity 0.86)
- Multi-Region Text-Driven Manipulation of Diffusion Imagery (AAAI 2024 · similarity 0.85)
- MVInpainter: Learning Multi-View Consistent Inpainting to Bridge 2D and 3D Editing (NeurIPS 2024 · similarity 0.85)
- AGILE3D: Attention Guided Interactive Multi-object 3D Segmentation (ICLR 2024 · similarity 0.85)
- TrajLoc: Trajectory-Attention Localization for Multi-Object Motion Control (NeurIPS 2026 · 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.
