When training the lmo dataset, all results are 0, and the translation and rotation errors are large.

[Fusica]

I encountered several issues when trying to train on the lom dataset and reproduce the author's results:

1. First, I found that in lmo2poet.py, the division method for the lmo dataset is quite different from what I downloaded from bop. The lmo downloaded from bop does not have the train_pbr and train_synt subfolders. I would like to ask which version of lmo the author used.

2. Then, when I loaded lmo_maskrcnn_checkpoint.pth.tar during training, the ADD was all 0. I want to know if it's because I made a mistake in the dataset generation step, or if there is something wrong with my training hyperparameters. I have referred to some of the hyperparameter settings provided in the author's link, but the results have not improved at all.

I sincerely hope the author can point out where my problem lies, as I am very eager to follow your work.

[Fusica]

I tried using the command python main.py --epochs 50 --batch_size 24 --output_dir outputs/202410231444 --resume weights/poet_lmo_maskrcnn.pth --eval to test the pth file you provided, and I got the following results. Does this match your results, and does this result prove that there is no problem with my dataset?

The angle error feels large, with 40°.

[tgjantos]

Hi @Fusica ,

I am really sorry that you faced these issues with the LM-O dataset. Regarding the download links I used the following, however I realize that they switched to huggingface for hosting their data:

train_pbr: https://huggingface.co/datasets/bop-benchmark/datasets/resolve/main/lm/lm_train_pbr.zip
train_synt: https://huggingface.co/datasets/bop-benchmark/datasets/resolve/main/lmo/lmo_train.zip
train_synt: https://huggingface.co/datasets/bop-benchmark/datasets/resolve/main/lm/lm_train.zip

I am not sure, whether the data changed over time.

Regarding the performance: I am aware of an issue that the LM-O performance is not reproducible. We have a network that is not performing better, but we still need some time before we can release it. I hope it does not hinder your work! I am really sorry.

Best,
Thomas