klingner

Hey there, Regarding 1. you could also try to train at a lower resolution, depending on what output resolution you need in the end. If 416/128 is enough for you,...

Hello! For this purpose you should probably take a look at the file ``loaders/segmentation/train.py``. Here, the different possible loaders for semantic segmentation are defined (in this case only cityscapes, but...

Hi, the basic_files.json is used to store information which color image belongs to which segmentation mask. Information on how to generate this file is available in this repository: https://github.com/ifnspaml/IFN_Dataloader/tree/master/dataloader/file_io. The...

Hi, the ground truth depth map is generated by the download_kitti.py script when downloading the KITTI data. Vizualization can be done analogous to the Cityscapes example you have posted here...

Hey there, this is an error I usually get, if the datasets' folder structure is either not correct or if some of the images are missing/broken. In this case the...

Hi! For training you would only need the image files. If you want to evaluate you also need ground truth. You will probably need to write your own data loading...