search

vincent-leguen / PhyDNet

Code for our CVPR 2020 paper "Disentangling Physical Dynamics from Unknown Factors for UnsupervisedVideo Prediction"
MIT License
179 stars 50 forks source link

Cannot find dataset #17

Open npmhung opened 3 years ago

npmhung commented 3 years ago

Hi,

I'm really interested in your paper and would like to reproduce the results on SST and Traffic BJ dataset. However, I couldn't find them because the links are dead.

Do you still keep a copy of them?

Could you send me through my email (nguyehu5@oregonstate.edu) or uploaded in google drive for downloading?

vincent-leguen commented 3 years ago

Hi @npmhung ,

You can find the Traffic BJ dataset from the MIM repository: https://github.com/Yunbo426/MIM I've just checked the Dropbox download link is correct.

And you can download SST data on this website: https://resources.marine.copernicus.eu/?option=com_csw&product_id=GLOBAL_ANALYSIS_FORECAST_PHY_001_024&view=details

Best Vincent

npmhung commented 3 years ago

Hi @vincent-leguen,

I have one more question about the Traffic BJ dataset.

In the Traffic BJ, each frame has 2 channels (inflow/outflow) of the traffic at a certain time. I wonder which channel you used to visualize in your paper?

Best, Hung

vincent-leguen commented 3 years ago

Hi @npmhung , we visualize the first channel in our figures. Best, Vincent

toddwyl commented 3 years ago

@vincent-leguen How to visualize the first channel ? I find the Traffic BJ dataset max_value is 1292. I have no idea about how to visualize it.

toddwyl commented 3 years ago

@vincent-leguen How to visualize the first channel ? I find the Traffic BJ dataset max_value is 1292. I have no idea about how to visualize it.

OK，I know that you use the colormap to convert gray into color. Is it using the cv2.COLORMAP_VIRIDIS ?

paapu88 commented 2 years ago

And you can download SST data on this website: https://resources.marine.copernicus.eu/?option=com_csw&product_id=GLOBAL_ANALYSIS_FORECAST_PHY_001_024&view=details

This links to a copernicus page with global forecast. Can you give any more data to reproduce the results in the paper? Which date, which variable, which prediction file, which lat lon area (they have 1/12 deg resolution and your input is 64x64)