Question regarding numpy arrays

VSainteuf / utae-paps

PyTorch implementation of U-TAE and PaPs for satellite image time series panoptic segmentation.

MIT License

140 stars 55 forks source link

Question regarding numpy arrays #25

Open fazelkhaha opened 1 year ago

fazelkhaha commented 1 year ago

Very well done job on the paper and the code!

I did have a question about the numpy arrays used for training, specifically the 'heatmap_xx.npy', 'zones_xx.npy', and 'instances_xx.npy' files. I've already found some information about the 'ParcelIDs_xx.npy' and 'target_xx.npy' files, but I was hoping you could provide me with some additional details on how the other files were created. Thank you!

VSainteuf commented 1 year ago

Hi @fazelkhaha , thanks!

the heatmap.npy is the ground truth for centerness, computed as described in the paper
the zones.npy is the mapping pixel -> object, you can check this issue for more details on the computation
the instances.npy contains the instance ids of each pixel of a patch. (each object gets a random instance_id and all the pixels of that object share this instance_id).