Daeun Lee
Daeun Lee
Hi, i occured **segmentation fault(core dumped) error** when import torch_sparse. Why this error occured? My version is .. ``` torch : 1.9.0+cu111 cuda : 11.4 ``` I installed torch-sparse using...
Hi, I have question about number of class-wise gt_sampling parameters. Is it **hyperparameter?** Also, if the class-wise sampling numbers are increased, is it correct that the GT of another scene...
Hi, thanks for your nice work! btw, is there any scene image in nuScenes dataset? like below in Kitti dataset. Best,
Hi, thank you for your nice work! I confirmed that the lane annotation of CULane consists of `4 (left-left, left-right, right-right)`. Therefore, it seems that CULane **has already been annotated**...
Hi, I wonder that why do this in the process of clustering features. https://github.com/NVIDIA/pix2pixHD/blob/5a2c87201c5957e2bf51d79b8acddb9cc1920b26/encode_features.py#L47 can anyone help me? plz
Hi, @Eromera thank u for your nice work! btw, in the `trained_models` directory, is `erfnet_pretrained.pth` used **ImageNet pretrained encoder**? and is it for segmentation to cityscapes dataset? How can I...
Hi, thank you for your nice work! btw, is there your code that generate `list6_train.txt` in TUSimple? I know that max lane in TUSimple is 5, but why above .txt...
Hi, @cardwing I'm a person who is not familar to matlab. **Can I get python version of `ERFNet-CULane-PyTorch/tools/prob2lines/main.m` codes?** I saw a https://github.com/cardwing/Codes-for-Lane-Detection/issues/83 issue that mentioned same approach to me.
Hi, thank you for your nice work! Im a person doing research on trajectory presentation and I have question about your **UCY datasets** preprocessing. Than, **Where did you collect raw...
Thank you for your nice work. Like in your paper fig.2 , How can I visualize diffpooling process? What method did you use? Best, Daeun Lee