Martin Hahner

I tried the hints from the closed issue "libsiftgpu.so failed to build." but they did not work for me.

thanks for the code @jfzhang95 I am also curious if you have some weights after training on Cityscapes as well. The highest mIoU I could achieve with the resnet backbone...

Hi Barza, the [splits](https://github.com/SysCV/LiDAR_snow_sim/tree/main/splits) given in this repo are "FOV3000strongest", so the "bad" frames are not in these lists anymore and these splits are the ones we used. > Can...

From wandb: In the paper:

> Thank you for replying. How did you get distance-wise mAP? Did you make that code public? see https://github.com/SysCV/LiDAR_snow_sim/issues/11

> I have marked in bold 2 AP results for moderate CAR. Which one did you report in your paper? > > INFO Car [email protected], 0.70, 0.70: bbox AP:62.0437, 59.4732,...

> I checked the frames in train_clear.txt. It contains some frames which have less than 3000 points in camera FOV (before augmenting snow). Did you only make 3000 points check...

> Also, you included FOG_AUGMENTATION_AFTER: False in all your configs for e.g. [here](https://github.com/MartinHahner/OpenPCDet/blob/main/tools/cfgs/dataset_configs/dense_dataset_snow_wet_coupled.yaml#L26) but in your [dense_dataset.py](https://github.com/MartinHahner/OpenPCDet/blob/0b8c57e321751b4712c233d008c8cbbed06419f5/pcdet/datasets/dense/dense_dataset.py#L906) you augment fog by just checking if 'FOG_AUGMENTATION_AFTER' is in the cfg or...

> Also, for me it has been impossible to obtain 40+ AP with 0.7 IoU threshold in 80 epochs for moderate Car, even when I train only on clear (without...

> Shouldn't [this](https://github.com/MartinHahner/OpenPCDet/blob/main/pcdet/datasets/dataset.py#L45) pass self.dataset_cfg.DATA_AUGMENTOR instead of self.dataset_cfg as augmentor_configs? Yes, you should change it (back) to `self.dataset_cfg.DATA_AUGMENTOR`. I played around with some changes in [DataAugmentor](https://github.com/open-mmlab/OpenPCDet/blob/master/pcdet/datasets/augmentor/data_augmentor.py#L9) for which I needed...