Daniel Seichter

The networks with ResNet50 backbone only serve as a reference for our approach. As shown in our experiments, the performance gain ist quite small while increasing inference time notably. However,...

The error comes from the global average pooling in the context module. The output of the global branch has no spatial resolution any more (see your error). If you are...

I am not sure whether I understand your question correctly. We are interested in labels for semantic segmentation only. So we do not need instance labels for semantic segmentation. For...

Even if each pixel is assigned to one of the known classes, the network might still be able to detect unknown objects with some height on the ground. During tests...

There is already an abstract base class for the datasets, just have a look at the dataset classes.

We are facing the same errors/warnings. It tells you that your version of TensorFlow is not compatible with the CUDA / cuDNN version installed by PyTorch. However, we do not...

We never faced this problem. The factor for random scaling is chosen between 1.0 and 1.4. So it's quite unlikely to pick a batch full of void. Which dataset and...

The problem comes from this os.path.join [here](https://github.com/TUI-NICR/ESANet/blob/main/src/datasets/sunrgbd/pytorch_dataset.py#L133), which adds the Windows path separator between the predefined path components. You can try to join the paths manually using "/" - python...

However, for qualitative results, you can still use [inference_dataset.py](https://github.com/TUI-NICR/ESANet/blob/main/inference_dataset.py) (see: [inference section](https://github.com/TUI-NICR/ESANet#dataset-inference) in our readme). If you want to store the evaluations results combining both [inference_dataset.py](https://github.com/TUI-NICR/ESANet/blob/main/inference_dataset.py) and [eval.py](https://github.com/TUI-NICR/ESANet/blob/main/eval.py) might be...

The error description is pretty short. Can you please provide some further information, i.e., environment (`conda list` / `pip list`), folder structure, executed command, and full error trace).