search

xyrui / PLRDiff

Unsupervised Hyperspectral Pansharpening via Low-rank Diffusion Model (Information Fusion 2024)
35 stars 6 forks source link

What is the meaning of the param 'PH' in test_single.py? #1

Open ToT-12136 opened 1 year ago

ToT-12136 commented 1 year ago

This param is multiplied by 'xhat_1' in the function _'p_sample_loopprogressive'.

What is the meaning of the param 'PH' in test_single.py?

PH = th.from_numpy(np.float32(data['H'])).unsqueeze(0).unsqueeze(-1)

In addition, can the author provide the datasets used in the actual test? The available links are not consistent with the treatment described in the paper

'Unsupervised Pansharpening via Low-rank Diffusion Model'

xyrui commented 3 months ago

"PH" means the spectral response vector "r" in Eq. (2).

I've uploaded all datasets used in our experiments for reproduce the results. Please find them in https://drive.google.com/drive/folders/161ExMLMeGTyckdyAGdGjN0hKe3ONemkD?usp=sharing.