search

TRAILab / CaDDN

Categorical Depth Distribution Network for Monocular 3D Object Detection (CVPR 2021 Oral)
Apache License 2.0
359 stars 62 forks source link

Problem the provided pretrained model #64

Closed YunzheWu-404 closed 2 years ago

YunzheWu-404 commented 2 years ago

Hi,

I tried to evaluate the pre-rained model provided on the GitHub homepage. My evaluation of this model on the KITTI 3D validation dataset shows a moderate 3D ap(R40) of 15.87. However, a moderate 3D ap (R40) of 16.31 is illustrated in table 3 and table 4 in your essay. Do you have any idea about it?

On the other hand, as I found on prevent issues, you mentioned that you have refactored your model. May I ask what specific changes have you made?

Best Nelson

codyreading commented 2 years ago

Hi Nelson,

The model provided on the homepage is a different run from the model reported in the paper. If you take a look at the results on the README you can see that the Moderate 3D AP is 16.07 with this model, which is similar to the results you report. Our results are very similar, where differences could include version differences or different hardware.

In terms of the refactoring or changes to the model since the paper, the model itself is the same. The refactoring was just some code refactoring, but all the learned components are identical. The small performance difference between the paper and the model reported here is just due to the variance between runs.

YunzheWu-404 commented 2 years ago

Thank you very much for your reply! It helps a lot.