Research map / Architectures and efficiency / CNN design
LAGConv: Local-Context Adaptive Convolution Kernels with Global Harmonic Bias for Pansharpening
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 standard convolution with LAGConv | pansharpening network · WV3 reduced resolution | SAM: 3.744 ± 1.22 -> 3.473 ± 1.19 | Helped |
| replace standard convolution with LAGConv | pansharpening network · GF2 reduced resolution | ERGAS: 1.00 ± 0.20 -> 0.91 ± 0.20 | Helped |
| replace standard convolution with LAGConv | pansharpening network · QB reduced resolution | ERGAS: 4.05 ± 0.26 -> 3.74 ± 0.29 | Helped |
| replace standard convolution with LAGConv | pansharpening network · WV3 full resolution | QNR: 0.9556 ± 0.031 -> 0.9637 ± 0.011 | Helped |
| add local-context adaptive kernels (ablation) | LAGConv pansharpening network · Tripoli WV3 | SAM: 4.2267 -> 4.0354 | Helped |
| add global harmonic bias to LCA kernels (ablation) | LAGConv pansharpening network · Tripoli WV3 | SAM: 4.0354 -> 3.9740 | Helped |
| use conventional bias instead of no bias (ablation) | conventional-kernel network · Tripoli WV3 | SAM: 4.2564 -> 4.3483 | Hurt |
| use GH bias instead of conventional bias (ablation) | conventional-kernel network · Tripoli WV3 | SAM: 4.3483 -> 4.2267 | Helped |
| use conventional bias instead of GH bias (ablation) | LCA-kernel network · Tripoli WV3 | SAM: 4.0354 -> 4.0264 | No clear effect |
| replace LAGConv with PAC | residual pansharpening network · WV3 reduced resolution | SAM: 4.10± 1.7 -> 3.47 ± 1.1 | Hurt |
| replace LAGConv with DDF | residual pansharpening network · WV3 reduced resolution | SAM: 3.87 ± 1.2 -> 3.47 ± 1.1 | Hurt |
| apply the proposed network to hyperspectral image super-resolution | CAVE dataset · CAVE | PSNR: 46 ± 2 -> 47 ± 3 | Helped |
Closest papers by meaning
- Content-Adaptive Non-Local Convolution for Remote Sensing Pansharpening (CVPR 2024 · similarity 0.92)
- PanAdapter: Two-Stage Fine-Tuning with Spatial-Spectral Priors Injecting for Pansharpening (AAAI 2025 · similarity 0.88)
- Adaptive Rectangular Convolution for Remote Sensing Pansharpening (CVPR 2025 · similarity 0.87)
- Pan-LUT: Efficient Pan-sharpening via Learnable Look-Up Tables (NeurIPS 2025 · similarity 0.87)
- A General Adaptive Dual-level Weighting Mechanism for Remote Sensing Pansharpening (CVPR 2025 · similarity 0.87)
- Dual-Granularity Semantic Guided Sparse Routing Diffusion Model for General Pansharpening (CVPR 2025 · similarity 0.87)
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.