Research map / Image generation / Editing and style transfer
SE360: Semantic Edit in 360° Panoramas via Hierarchical Data Construction
Haoyi Zhong, Fang-Lue Zhang, Andrew Chalmers, Taehyun Rhee
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 |
|---|---|---|---|
| fine-tune on SE360-HF after SE360-Base pre-training (two-stage) (ablation) | SE360 · PanoEval object addition | CSpers (CLIP score): 27.965 -> 29.354 | Improved |
| fine-tune on SE360-HF after SE360-Base for object removal (ablation) | SE360 · PanoEval object removal | FAED: 0.498 -> 0.314 | Improved |
| train only on SE360-HF instead of two-stage (ablation) | SE360 · PanoEval object removal | PSNR: 26.394 -> 19.419 | Hurt |
| train only on SE360-HF instead of two-stage (ablation) | SE360 · PanoEval object addition | LPIPS: 0.076 -> 0.171 | Hurt |
| fine-tune on SE360-HF (remove artifact overfitting) when evaluated on LaMa-artifact-containing data (ablation) | SE360 · PanoEval object addition | CSpers (CLIP score): 29.377 -> 29.354 | Hurt |
| train only on SE360-Base (LaMa-erased data) instead of adding HF refinement (ablation) | SE360 · PanoEval object removal | CS-Nopers: 73.143 -> 73.055 | Hurt |
| SE360 versus Step1X-Edit on removal reconstruction metrics | SE360 · PanoEval object removal | PSNR: SE360 26.394 vs Step1X-Edit 28.471 | Hurt |
| SE360 versus Flux.1 Kontext dev on addition semantic score | SE360 · PanoEval object addition | CSpers (CLIP score): 29.295 -> 29.354 | No clear effect |
| SE360 versus top-performing models in user study | SE360 · User study consistency | Likert score: 2.91 -> 4.47 (vs Flux.1 Kontext dev) | 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.
- OmniGen — OmniGen: Unified Image Generation (CVPR 2025) · 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
- HORIZON: High-Resolution Semantically Controlled Panorama Synthesis (AAAI 2024 · similarity 0.87)
- CubeDiff: Repurposing Diffusion-Based Image Models for Panorama Generation (ICLR 2025 · similarity 0.86)
- PanoWorld: Towards Spatial Supersensing in 360◦ Panorama World (NeurIPS 2026 · similarity 0.85)
- EditSplat: Multi-View Fusion and Attention-Guided Optimization for View-Consistent 3D Scene Editing with 3D Gaussian Splatting (CVPR 2025 · similarity 0.84)
- DiffPano: Scalable and Consistent Text to Panorama Generation with Spherical Epipolar-Aware Diffusion (NeurIPS 2024 · similarity 0.84)
- Taming Stable Diffusion for Text to 360 Panorama Image Generation (CVPR 2024 · 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.
