Yining Shi
Yining Shi
I am looking for tutorials of camera-only challenge, about how to load 3D and 2D annotations introduced by new challenge, e.g 'camera_synced_box' and 'most_visible_camera_name'. Could you please upload some tutorials...
Hi, I am building the ros version of dogm, and at the last stage I failed possibly because of some dependencies issues. my error is like follows: /usr/bin/ld: /usr/local/lib/libdogm.a(dogm.cu.o): relocation...
I have implemented your training process on nuScenes dataset. I used your default settings. My environment is like 4x3090, but the error is like "CUDA out of memory". How I...
System: 4xRTX3090. Training scripts: Default training scripts: python -u train_lseg.py --dataset ade20k --data_path ./datasets --batch_size 1 --exp_name lseg_ade20k_l16 \ --base_lr 0.004 --weight_decay 1e-4 --no-scaleinv --max_epochs 200 --widehead --accumulate_grad_batches 2 --backbone...
Thanks for your great work. I have some doubt reproducing your results about data preparation. The data structure puzzles me because they are not the default structure of KITTI, nuScenes...
System: Ubuntu18.04LTS PCL1.9.1 I am compiling the source file in the tutorial pcl-pcl-1.9.1/doc/tutorials/content/sources/point_cloud_compression. By default it works well. The readthedocs said the compression supports any point types, but when I...
I find that when using the SensorDataloader, some lidar timestamps are not synchronized to seven ring cameras. Some only give six synchronized imageries. Is the tolerance of synchronization too strict?
Thanks for your great work. I replace the anchor3d head with centerhead, but the algorithm does not converge, have you tested masked SST with CenterHead?
Thanks for your great work. May you provide some examples on images where width is not equal to height. Thanks.