how to resolve

[ToTensor]

Traceback (most recent call last):
File "tools/train.py", line 263, in
main()
File "tools/train.py", line 259, in main
meta=meta)
File "/home/ly/tr3d-main/mmdet3d/apis/train.py", line 351, in train_model
meta=meta)
File "/home/ly/tr3d-main/mmdet3d/apis/train.py", line 319, in train_detector
runner.run(data_loaders, cfg.workflow)
File "/home/ly/anaconda3/envs/tr3d-main/lib/python3.7/site-packages/mmcv/runner/epoch_based_runner.py", line 136, in run
epoch_runner(data_loaders[i], **kwargs)
File "/home/ly/anaconda3/envs/tr3d-main/lib/python3.7/site-packages/mmcv/runner/epoch_based_runner.py", line 49, in train
for i, data_batch in enumerate(self.data_loader):
File "/home/ly/anaconda3/envs/tr3d-main/lib/python3.7/site-packages/torch/utils/data/dataloader.py", line 521, in next
data = self._next_data()
File "/home/ly/anaconda3/envs/tr3d-main/lib/python3.7/site-packages/torch/utils/data/dataloader.py", line 1203, in _next_data
return self._process_data(data)
File "/home/ly/anaconda3/envs/tr3d-main/lib/python3.7/site-packages/torch/utils/data/dataloader.py", line 1229, in _process_data
data.reraise()
File "/home/ly/anaconda3/envs/tr3d-main/lib/python3.7/site-packages/torch/_utils.py", line 434, in reraise
raise exception
ValueError: Caught ValueError in DataLoader worker process 0.
Original Traceback (most recent call last):
File "/home/ly/anaconda3/envs/tr3d-main/lib/python3.7/site-packages/torch/utils/data/_utils/worker.py", line 287, in _worker_loop
data = fetcher.fetch(index)
File "/home/ly/anaconda3/envs/tr3d-main/lib/python3.7/site-packages/torch/utils/data/_utils/fetch.py", line 49, in fetch
data = [self.dataset[idx] for idx in possibly_batched_index]
File "/home/ly/anaconda3/envs/tr3d-main/lib/python3.7/site-packages/torch/utils/data/_utils/fetch.py", line 49, in
data = [self.dataset[idx] for idx in possibly_batched_index]
File "/home/ly/anaconda3/envs/tr3d-main/lib/python3.7/site-packages/mmdet/datasets/dataset_wrappers.py", line 178, in getitem
return self.dataset[idx % self._ori_len]
File "/home/ly/tr3d-main/mmdet3d/datasets/custom_3d.py", line 435, in getitem
data = self.prepare_train_data(idx)
File "/home/ly/tr3d-main/mmdet3d/datasets/custom_3d.py", line 225, in prepare_train_data
input_dict = self.get_data_info(index)
File "/home/ly/tr3d-main/mmdet3d/datasets/sunrgbd_dataset.py", line 112, in get_data_info
depth2img = calib['K'] @ rt_mat
ValueError: matmul: Input operand 1 has a mismatch in its core dimension 0, with gufunc signature (n?,k),(k,m?)->(n?,m?) (size 3 is different from 9)

[filaPro]

Can you print the shapes of calib['K'] and rt_mat? I think they both should be (3, 3), as we save them here.

[filaPro]

Why do I fail to scan the scan dataset when running Python batchreload_scannet_data.py

can you please share the full error log?

[ToTensor]

（9，9）

I have not solved this problem at present, and the shape of both of them is 9

[filaPro]

may you try to reshape 9 to (3, 3)?

btw did you run this script? are both matrices (3, 3) after it?

[ToTensor]

python batch_load_scannet_data.py
--------------------begin
2024-02-29 19:04:14.024408
scene0000_00
Failed export scan: scene0000_00
--------------------done
--------------------begin
2024-02-29 19:04:14.831180
scene0000_01
Failed export scan: scene0000_01
--------------------done
This error occurred when I was processing the scannetv2 dataset

[filaPro]

Looks like the real error is hidden by this exception. Can you please share the error trace at this line?

[ToTensor]

FileNotFoundError: [Errno 2] No such file or directory: 'scans/scene0000_00/scene0000_00.txt'
` # try:
export_one_scan(scan_name, output_filename_prefix, max_num_point,
label_map_file, scannet_dir, test_mode)

    #except Exception:
        #print(f'Failed export scan: {scan_name}')
    print('-' * 20 + 'done')`

When I remove the thrown exception

[filaPro]

Have you downloaded the scannet dataset and put it to this scans/ directory? The first item in the instruction here.

[ToTensor]

。thanks！The problem has been solved, btw. May I ask how to display the correct boxes in the visualization of the obj file

[filaPro]

We save both gt and predictied boxes for visualization with this function. Resulting .obj files can be opened with the meshlab together with point cloud .obj file.

[ToTensor]

Why the gt visualized by the sunrgbd data set is slanted and does not completely wrap the real object. And pred.obj visualizes a lot of boxes, how can I show the box that is closest to gt?

[filaPro]

To reduce the number of predicted boxes please follow our advice about score_thr in the readme.

Gt boxes looks pretty much fine. May be you can try it on scene 000967 from SUN RGB-D, to be sure that it is the similar with gt visualization in our paper.