代码拉取完成,页面将自动刷新
A Simple PointPillars PyTorch Implenmentation for 3D Lidar(KITTI) Detection.
cd ops
python3 setup.py develop
Download
Download point cloud(29GB), images(12 GB), calibration files(16 MB)和labels(5 MB)。 Format the datasets as follows:
kitti
|- ImageSets
|- train.txt
|- val.txt
|- test.txt
|- trainval.txt
|- training
|- calib (#7481 .txt)
|- image_2 (#7481 .png)
|- label_2 (#7481 .txt)
|- velodyne (#7481 .bin)
|- testing
|- calib (#7518 .txt)
|- image_2 (#7518 .png)
|- velodyne (#7418 .bin)
The train.txt、val.txt、test.txt and trainval.txt you can get from:
wget https://raw.githubusercontent.com/traveller59/second.pytorch/master/second/data/ImageSets/test.txt
wget https://raw.githubusercontent.com/traveller59/second.pytorch/master/second/data/ImageSets/train.txt
wget https://raw.githubusercontent.com/traveller59/second.pytorch/master/second/data/ImageSets/val.txt
wget https://raw.githubusercontent.com/traveller59/second.pytorch/master/second/data/ImageSets/trainval.txt
Pre-process KITTI datasets First
ln -s path/to/kitti/ImageSets ./dataset
python3 pre_process_kitti.py --data_root your_path_to_kitti
Now, we have datasets as follows:
kitti
|- training
|- calib (#7481 .txt)
|- image_2 (#7481 .png)
|- label_2 (#7481 .txt)
|- velodyne (#7481 .bin)
|- velodyne_reduced (#7481 .bin)
|- testing
|- calib (#7518 .txt)
|- image_2 (#7518 .png)
|- velodyne (#7518 .bin)
|- velodyne_reduced (#7518 .bin)
|- kitti_gt_database (# 19700 .bin)
|- kitti_infos_train.pkl
|- kitti_infos_val.pkl
|- kitti_infos_trainval.pkl
|- kitti_infos_test.pkl
|- kitti_dbinfos_train.pkl
python3 train.py --data_root your_path_to_kitti
python3 -m torch.distributed.launch --nproc_per_node 8 train_dist.py --data_root your_path_to_kitti
python3 evaluate.py --ckpt pretrained/your_weights.pth --data_root your_path_to_kitti
# 1. infer and visualize point cloud detection
python3 test.py --ckpt pretrained/your_weights.pth --pc_path your_pc_path
# 2. infer and visualize point cloud detection and gound truth.
python3 test.py --ckpt pretrained/your_weights.pth --pc_path your_pc_path --calib_path your_calib_path --gt_path your_gt_path
# 3. infer and visualize point cloud & image detection
python3 test.py --ckpt pretrained/your_weights.pth --pc_path your_pc_path --calib_path your_calib_path --img_path your_img_path
e.g. [infer on val set 000134]
python3 test.py --ckpt pretrained/your_weights.pth --pc_path /home/lifa/data/KITTI/training/velodyne_reduced/000134.bin
or
python3 test.py --ckpt pretrained/your_weights.pth --pc_path data/kitti/training/velodyne_reduced/000134.bin --calib_path data/kitti/training/calib/000134.txt --img_path data/kitti/training/image_2/000134.png --gt_path data/kitti/training/label_2/000134.txt
Fast Encoders for Object Detection from Point Clouds](https://arxiv.org/abs/1812.05784)
此处可能存在不合适展示的内容,页面不予展示。您可通过相关编辑功能自查并修改。
如您确认内容无涉及 不当用语 / 纯广告导流 / 暴力 / 低俗色情 / 侵权 / 盗版 / 虚假 / 无价值内容或违法国家有关法律法规的内容,可点击提交进行申诉,我们将尽快为您处理。