Dimitris Mallios
Dimitris Mallios
I am interested too. Has anyone tried to do that yet?
Is anyone working on this? I would be happy to take it and do a PR
I noticed the same issue too. I think these are mislabeled...
I have the same question as well! If someone from the team could provide any help, that would be great!
@whatevercyh I have an implementation here https://github.com/dimimal/semantics_segmentation_of_urban_environments/blob/master/trainCRF.py in crfRNN function with (512,512,3) as you refer
Hello, Well, I remember that I saw these parameters on a paper and I was experimenting with the model correspondence on theta variations. Obviously, you can set theta_b = 3...
I am not sure if this has to do with the script. Can you give me the command you are entering to the terminal?
This is a problem in denseExtraction. Use this denseExtraction [here](https://github.com/dimimal/semantics_segmentation_of_urban_environments/blob/devel/denseExtraction.py) instead. Change the rows and columns and patchsize. I am working onto a new data generator pipeline right now and...
Remove the CRF flag and tell me if the same error occurs. Ensure that you have the necessary lib versions from requirements.txt
It is obvious that your data were not transformed correctly. Ensure that you're numpy files are correct. Try to load on a different script your labels, the correct shape of...