cvlab-yonsei/hvpr

An official implementation of "HVPR: Hybrid Voxel-Point Representation for Single-stage 3D Object Detection" (CVPR 2021) in PyTorch.

hvpr's Issues

I have encountered the following problems. How can I solve them

(HVPR) libing@linux:~/newstorage/3Dobjectdetector/HVPR/tools$ python train.py --cfg_file ./cfgs/kitti_models/hvpr.yaml
Traceback (most recent call last):
File "train.py", line 6, in
from test import repeat_eval_ckpt
File "/home/libing/newstorage/3Dobjectdetector/HVPR/tools/test.py", line 13, in
from eval_utils import eval_utils
File "/home/libing/newstorage/3Dobjectdetector/HVPR/tools/eval_utils/eval_utils.py", line 8, in
from pcdet.models import load_data_to_gpu
File "/home/libing/newstorage/3Dobjectdetector/HVPR/pcdet/models/init.py", line 6, in
from .detectors import build_detector
File "/home/libing/newstorage/3Dobjectdetector/HVPR/pcdet/models/detectors/init.py", line 1, in
from .detector3d_template import Detector3DTemplate
File "/home/libing/newstorage/3Dobjectdetector/HVPR/pcdet/models/detectors/detector3d_template.py", line 7, in
from .. import backbones_2d, backbones_3d, dense_heads, roi_heads, transformer
File "/home/libing/newstorage/3Dobjectdetector/HVPR/pcdet/models/backbones_3d/init.py", line 1, in
from .pointnet2_backbone import PointNet2Backbone, PointNet2MSG, PointNet2MSG_NOFP
File "/home/libing/newstorage/3Dobjectdetector/HVPR/pcdet/models/backbones_3d/pointnet2_backbone.py", line 5, in
from ...ops.pointnet2.pointnet2_stack import pointnet2_modules as pointnet2_modules_stack
File "/home/libing/newstorage/3Dobjectdetector/HVPR/pcdet/ops/pointnet2/pointnet2_stack/pointnet2_modules.py", line 7, in
from . import pointnet2_utils
File "/home/libing/newstorage/3Dobjectdetector/HVPR/pcdet/ops/pointnet2/pointnet2_stack/pointnet2_utils.py", line 5, in
from . import pointnet2_stack_cuda as pointnet2
ImportError: /home/libing/newstorage/3Dobjectdetector/HVPR/pcdet/ops/pointnet2/pointnet2_stack/pointnet2_stack_cuda.cpython-37m-x86_64-linux-gnu.so: undefined symbol: Z25voxel_query_wrapper_stackiiiiifiiiN2at6TensorES0_S0_S0_S0

How long to get fully trained?

Hi, I'd like to know what GPUs you used and how long did it take to train on each dataset.

Cannot find scripts directory

There's no scripts/ and ops/ directory in this project. Would you please push the whole project to the remote?

Your project can not run！

x86_64-linux-gnu-gcc: error: pcdet/ops/iou3d_nms/src/iou3d_cpu.cpp: 没有那个文件或目录

when I run "python3 setup.py develop" ,I got this error :

building 'pcdet.ops.iou3d_nms.iou3d_nms_cuda' extension x86_64-linux-gnu-gcc -pthread -DNDEBUG -g -fwrapv -O2 -Wall -g -fstack-protector-strong -Wformat -Werror=format-security -Wdate-time -D_FORTIFY_SOURCE=2 -fPIC -I/home/user/.local/lib/python3.6/site-packages/torch/include -I/home/user/.local/lib/python3.6/site-packages/torch/include/torch/csrc/api/include -I/home/user/.local/lib/python3.6/site-packages/torch/include/TH -I/home/user/.local/lib/python3.6/site-packages/torch/include/THC -I/usr/local/cuda/include -I/usr/include/python3.6m -c pcdet/ops/iou3d_nms/src/iou3d_cpu.cpp -o build/temp.linux-x86_64-3.6/pcdet/ops/iou3d_nms/src/iou3d_cpu.o -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_EXTENSION_NAME=iou3d_nms_cuda -D_GLIBCXX_USE_CXX11_ABI=0 -std=c++14 x86_64-linux-gnu-gcc: error: pcdet/ops/iou3d_nms/src/iou3d_cpu.cpp: 没有那个文件或目录 x86_64-linux-gnu-gcc: fatal error: no input files compilation terminated. error: command 'x86_64-linux-gnu-gcc' failed with exit status 1

Runnable code

The projiect is not complete

Thanks for your work, but the project loss some file ,Could you provide the Complete project?

Is this program complete? It feels like a lot of files are missing

Is this program complete? It feels like a lot of files are missing

Functions Missing

Hi,

I believe "SpatialAttention()" is missing from the 2d backbone.py. I got the below error:
NameError: name 'SpatialAttention' is not defined

It would be helpful to get the updated file.

Thanks and Regards
Niranjan Ravi.

Which commit does this pcdet based?

Hi, could you tell which commit of OpenPCDet does this project referred?

Asking about the point cloud range?

Thanks for your work.
The default config of point cloud range in the code is [0, -19.84, -2.5, 47.36, 19.84, 0.5] (code here). It is different from the point cloud range that was mentioned in the paper ([(0, 70.4),(−40, 40),(−3, 1)] meters).
What was the config that you used for the paper?
From my point of view, it's not fair to compare your results to other methods if the models are trained with different settings.

ModuleNotFoundError: No module named 'pcdet.datasets.nuscenes'

File "/home/user/HVPR/pcdet/datasets/init.py", line 9, in
from .nuscenes.nuscenes_dataset import NuScenesDataset
ModuleNotFoundError: No module named 'pcdet.datasets.nuscenes'

I got this error when I try to run 'python3 -m pcdet.datasets.kitti.kitti_dataset create_kitti_infos tools/cfgs/dataset_configs/kitti_dataset.yaml'

The project is uncomplete!!!!!!!!!!!!

The project is uncomplete!!!!!!!!!!!!