starrah/dmh-net

Code for 3D Room Layout Estimation from a Cubemap of Panorama Image via Deep Manhattan Hough Transform (ECCV 2022)

dmh-net's Issues

how to produce 3D mesh

how to produce 3D mesh, there is any code about it?

Error when training on other dataset

Hi, author. Thanks for your nice work!
I tried to train on the ZInD dataset, but I got some error message.
I don't change any code you provided.
By the way, should I use data augmentation if I have 20077 train set size?

• Environment
  Ubuntu 20.04.5 LTS

• Command

python train.py --cfg_file cfgs/zind.yaml --id zind -b 8 --train_root /home/user/CBS/DMH-Net/data/zind_layout/train --valid_root_dir /home/user/CBS/DMH-Net/data/zind_layout/valid

• Error Message

num_workers: 8
batch_size: 8
train_set_size: 20077
LR 0.000100
Train ep1:   0%|                                       | 0/2510 [00:00<?, ?it/s]
Epoch:   0%|                                             | 0/75 [00:00<?, ?ep/s]
Traceback (most recent call last):
  File "train.py", line 320, in <module>
    input = next(iterator_train)
  File "/home/user/anaconda3/envs/dmhnet/lib/python3.7/site-packages/torch/utils/data/dataloader.py", line 521, in __next__
    data = self._next_data()
  File "/home/user/anaconda3/envs/dmhnet/lib/python3.7/site-packages/torch/utils/data/dataloader.py", line 1203, in _next_data
    return self._process_data(data)
  File "/home/user/anaconda3/envs/dmhnet/lib/python3.7/site-packages/torch/utils/data/dataloader.py", line 1229, in _process_data
    data.reraise()
  File "/home/user/anaconda3/envs/dmhnet/lib/python3.7/site-packages/torch/_utils.py", line 425, in reraise
    raise self.exc_type(msg)
KeyError: Caught KeyError in DataLoader worker process 0.
Original Traceback (most recent call last):
  File "/home/user/anaconda3/envs/dmhnet/lib/python3.7/site-packages/torch/utils/data/_utils/worker.py", line 287, in _worker_loop
    data = fetcher.fetch(index)
  File "/home/user/anaconda3/envs/dmhnet/lib/python3.7/site-packages/torch/utils/data/_utils/fetch.py", line 44, in fetch
    data = [self.dataset[idx] for idx in possibly_batched_index]
  File "/home/user/anaconda3/envs/dmhnet/lib/python3.7/site-packages/torch/utils/data/_utils/fetch.py", line 44, in <listcomp>
    data = [self.dataset[idx] for idx in possibly_batched_index]
  File "/home/user/Laurence/DMH-Net/perspective_dataset.py", line 101, in __getitem__
    return self.getItem(self.img_fnames[idx], self.cors[idx])
  File "/home/user/Laurence/DMH-Net/perspective_dataset.py", line 286, in getItem
    for line in d["lines"]:
KeyError: 'lines'

Inferene on my own dataset

Thank you for your nice work!

I try to run your code on my own erp images.

It seems ok to run your inference model with only images.

However, your dataloader code requires '.txt' files about corners.

I think it is needed in only training situation, not inference status.

If so, how I can run your code on my dataset?
If not, how I can generate '.txt' files?

Pre-trained Model

Hello,
Is there a link to access your pre-trained model?

Unexpected segmentation fault encountered in worker.

Thank you for the nice work, but I face the unexpected segmentation fault encountered in worker even with num_workers set to 0. Error message are listed below. I use NVIDIA Tesla V100. Do you have any idea? Thanks.

batch_size: 8
train_set_size: 403
LR 0.000100
Epoch:   0%|                                                                                                       | 0/75 [00:00<?, ?ep/sERROR: Unexpected segmentation fault encountered in worker.                                                         | 0/51 [00:00<?, ?it/s]
ERROR: Unexpected segmentation fault encountered in worker.
ERROR: Unexpected segmentation fault encountered in worker.
ERROR: Unexpected segmentation fault encountered in worker.
ERROR: Unexpected segmentation fault encountered in worker.
ERROR: Unexpected segmentation fault encountered in worker.
Train ep1:   0%|                                                                                                   | 0/51 [00:00<?, ?it/s]
Epoch:   0%|                                                                                                       | 0/75 [00:01<?, ?ep/s]
Traceback (most recent call last):
  File "/home/u111061517/.local/lib/python3.7/site-packages/torch/utils/data/dataloader.py", line 761, in _try_get_data
    data = self._data_queue.get(timeout=timeout)
  File "/home/u111061517/miniconda3/envs/dmhnet/lib/python3.7/queue.py", line 179, in get
    self.not_empty.wait(remaining)
  File "/home/u111061517/miniconda3/envs/dmhnet/lib/python3.7/threading.py", line 300, in wait
ERROR: Unexpected segmentation fault encountered in worker.
ERROR: Unexpected segmentation fault encountered in worker.
    gotit = waiter.acquire(True, timeout)
  File "/home/u111061517/.local/lib/python3.7/site-packages/torch/utils/data/_utils/signal_handling.py", line 66, in handler
    _error_if_any_worker_fails()
RuntimeError: DataLoader worker (pid 270398) is killed by signal: Segmentation fault. 

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "train.py", line 321, in <module>
    input = next(iterator_train)
  File "/home/u111061517/.local/lib/python3.7/site-packages/torch/utils/data/dataloader.py", line 345, in __next__
    data = self._next_data()
  File "/home/u111061517/.local/lib/python3.7/site-packages/torch/utils/data/dataloader.py", line 841, in _next_data
    idx, data = self._get_data()
  File "/home/u111061517/.local/lib/python3.7/site-packages/torch/utils/data/dataloader.py", line 798, in _get_data
    success, data = self._try_get_data()
  File "/home/u111061517/.local/lib/python3.7/site-packages/torch/utils/data/dataloader.py", line 774, in _try_get_data
    raise RuntimeError('DataLoader worker (pid(s) {}) exited unexpectedly'.format(pids_str))
RuntimeError: DataLoader worker (pid(s) 270395, 270398, 270405, 270406, 270407) exited unexpectedly
Segmentation fault (core dumped)

The following error message occurred when I set num_workers to 0

num_workers: 8
batch_size: 8
train_set_size: 403
LR 0.000100
Epoch:   0%|               | 0/75 [00:00<?, ?ep/sSegmentation fault (core dumped)     
| 0/51 [00:00<?, ?it/s]

Inference and 3D visualization code for custom image

Thank you for your excellent work. Is it possible to release the inference code for custom images that don't come with a JSON or TXT file, just the image itself? Since the dataset isn't publicly available, it's a bit challenging to determine how the custom image's structure should be for evaluation to work as a inference code. same issue for visualization in 3d .