SuperLine3D
SuperLine3D copied to clipboard
zxrzju
[ECCV 2022] SuperLine3D: Self-supervised Line Segmentation and Description for LiDAR Point Cloud
兄弟，你发的谷歌网盘数据集，限速一小时下载下来，超过之后token就会失效，下载失败，可以放个百度网盘的链接嘛。国外网限速呀，痛苦
Hello! Thanks for open-sourcing this work. When I run: `python train_synthetic_data.py --load_folder scripts/synthetic_data/line_pole_noise/` The observation is that the GPU is not actually used (only 300~MB occupied), but the CPU-util increases...
Great work! I am wondering if you have any plans to release your trained model. It will help anyone who is interested in your work to get a better understanding...
tensorflow.python.framework.errors_impl.InvalidArgumentError: indices[24] = 25 is not in [0, 25) [[node tower_0/while_1/boolean_mask/GatherV2 (defined at home/zhaoqian/anaconda3/envs/tf-gpu/lib/python3.7/site-packages/tensorflow_core/python/framework/ops.py:1748) ]]
Thank you for your excellent contribution in point cloud alignment based on self-supervised line primitives, I have encountered a problem in the process of reproducing your project, the get_model() function...
Thank you for your excellent contribution in point cloud alignment based on self-supervised line primitives, I have encountered a problem in the process of reproducing your project, the get_model() function...
Thank you for your excellent contribution in point cloud alignment based on self-supervised line primitives, I have encountered a problem in the process of reproducing your project, the get_model() function...