zhyever

Sorry for the late. Download the [pkl](https://drive.google.com/file/d/19W4HGKBqmfxeKyTNVS6WTnkZd5fLTGgg/view?usp=sharing) and the [zipped gt_database](https://drive.google.com/file/d/1pP63GzZP_cg-mLYQmEmDDW1mwdlZW-cF/view?usp=sharing). Rename the pkl file to `kitti_dbinfos_train.pkl` and put it under your data folder. Unzip the .zip file, rename the...

Set this flag in your config file instead of passing it by the shell. You can add a line of `find_unused_parameters=True` in your config file.

That's something related to the build of mmdet3d (in this repo, SimIPU). Refer to [Issue](https://github.com/open-mmlab/mmdetection3d/issues/513) for more information.

Maybe not. Remember that we need to adopt a matching algorithm to get positive pairs in the intro branch. There could be mistakes in matching. Also, the positive pairs do...

Please refer to the 3D detection log presented [here](https://github.com/zhyever/SimIPU/blob/main/resources/logs/moca_simipu_kitti.txt). The model performance in the last several epochs is consistently better than the baseline w/o pertaining. No other tricks were adopted...

Could you please try other SimIPU pre-trained models, so that I can ensure that it is my mistake uploading wrong models.

:D Hi, I would like to ask if there is any update. Remember when pre-training, there are two encoders actually (img, point cloud). So when loading the parameters, the point...

Hi, some discussions can be found: https://github.com/zhyever/PatchFusion/issues/24 For me, the metric depth estimator provides absolute depth, which is somehow useful in certain practical cases. It would be better to investigate...

Tbh, I have no idea about this issue. Anyone has a similar case and can help?

While I don't know what is going on, one hack implementation is as simple as replacing this `config.coarse_branch.pretrained_resource = None` with `config.coarse_branch['pretrained_resource'] = None`. L78 should be similar. I'm not...