Research map / Image generation / Image restoration
Semantic-Adaptive Diffusion for Dynamic Spatiotemporal Fusion
Jinsong Zhang, Ying Qu, Yuan Liao, Hairong Qi, Zhenzhou Shao
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 |
|---|---|---|---|
| replace RDDM with the Taylor-inspired decoder (ablation) | SA-STF · LGC dataset | SAM: 0.0683 -> 0.0652 | Improved |
| Taylor decoder alone without TFA and SAF (ablation) | SA-STF · LGC dataset | CC: 0.9194 -> 0.9194 | No clear effect |
| add temporal feature alignment (TFA) (ablation) | Taylor-inspired framework · LGC dataset | SAM: 0.0652 -> 0.0621 | Improved |
| add semantic-adaptive fusion (SAF) (ablation) | Taylor-inspired framework · LGC dataset | SAM: 0.0652 -> 0.0562 | Improved |
| add SAF (ablation) | Taylor-inspired framework · LGC dataset | ERGAS: 0.8486 -> 0.7230 | Improved |
| combine TFA and SAF in the full model (ablation) | SA-STF · LGC dataset | SAM: 0.0562 -> 0.0531 | Improved |
| add the diffusion process on top of TFA+SAF (ablation) | SA-STF · LGC dataset | ERGAS: 0.7045 -> 0.6751 | Improved |
| TFA+SAF without diffusion (ablation) | SA-STF · LGC dataset | CC: 0.9306 vs 0.9360 | No clear effect |
| use SA-STF instead of ESTARFM | STF methods · LGC dataset | RMSE: 0.0227 -> 0.0169 | Improved |
| use deep learning methods instead of linear ESTARFM on short time spans | STFMamba, SwinSTFM, STFDiff · CIA dataset | CC: 0.8823 -> 0.8172 (STFMamba) | Hurt |
Closest papers by meaning
- DiffusionSat: A Generative Foundation Model for Satellite Imagery (ICLR 2024 · similarity 0.87)
- SSDiff: Spatial-spectral Integrated Diffusion Model for Remote Sensing Pansharpening (NeurIPS 2024 · similarity 0.86)
- Efficient Representation Learning of Satellite Image Time Series and Their Fusion for Spatiotemporal Applications (AAAI 2024 · similarity 0.86)
- Causal Deciphering and Inpainting in Spatio-Temporal Dynamics via Diffusion Model (NeurIPS 2024 · similarity 0.86)
- Building Bridges across Spatial and Temporal Resolutions: Reference-Based Super-Resolution via Change Priors and Conditional Diffusion Model (CVPR 2024 · similarity 0.86)
- MMM-RS: A Multi-modal, Multi-GSD, Multi-scene Remote Sensing Dataset and Benchmark for Text-to-Image Generation (NeurIPS 2024 · 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.
