speed8928

I lowered the Theano version 7.x then it worked.

That would be a good idea Rain. I always find the inconsistent DSM patch from one to another although I train the model with 250 pixel overlap throughout the whole...

Hey, I used NDSM(lastools) from Potsdam while using DSM from Vaihingen for training. The path is in csv file. You just need to download the datasets and split into small...

For training Vaihingen dataset and Potsdam dataset, I converted image to 16 bit to keep precision and convert back to meters. While loading the data, you need to be careful...

I suggest you read the document for both of the datasets to understand what is their value stands. Multiplying 100000 is the way to convert to 16 bit as I...

> I used your released code for training and did not use any preprocessing on these two datasets (Vaihingen dataset and Potsdam dataset) as mentioned in the paper. I currently...

Hi opee Have a peep in nyu_transform.py the line:294 depth = self.to_tensor(depth)/100000 When images converted to tensor I divided into 100000. If you want to try to train your own...

Try a = np.array(250*1000).astype(np.uint16) a = a / 100000 yes, you need to normalize the DSM by subtracting DTM.

I solved it by reinstall opencv 2.4.9 form https://github.com/lakehanne/opencv-2.4.9 and specify all dependencies with this opencv and follow the installation guide with this page https://github.com/HTLife/Kintinuous-installation-guide/blob/master/Kintinuous-installation-guide-16.04.md

Hi maybe just don't use ssim for evaluation. It has been depreciated in the newer pytorch version On Wed, 26 Jul 2023 at 12:27, MavrahShahid ***@***.***> wrote: > Thank you...