Accepted by TGRS 2024

QuickStart

1. download checkpoint from openClip
2. update config.general_finetune.json
3. python general_sharpening_joint_random_batch_finetune.py

Cite

Y. Xing, L. Qu, S. Zhang, J. Feng, X. Zhang and Y. Zhang, "Empower Generalizability for Pansharpening Through Text-Modulated Diffusion Model," in IEEE Transactions on Geoscience and Remote Sensing, vol. 62, pp. 1-12, 2024, Art no. 5633812, doi: 10.1109/TGRS.2024.3434431. keywords: {Satellites;Pansharpening;Diffusion models;Noise reduction;Training;Data models;Three-dimensional displays;Denoising diffusion probabilistic model (DPM);text-modulated model;unified pansharpening},