yancie-yjr / dbq-ssd Goto Github PK
View Code? Open in Web Editor NEWThe official implementation of the paper DBQ-SSD: Dynamic Ball Query for Efficient 3D Object Detection (ICLR 2023)
License: Apache License 2.0
The official implementation of the paper DBQ-SSD: Dynamic Ball Query for Efficient 3D Object Detection (ICLR 2023)
License: Apache License 2.0
Could you give any advices to me? I think this kind of gate-based dynamic networks seem hard to converge.
Car AP@0.70, 0.70, 0.70:
bbox AP:66.2887, 50.9406, 50.1121
bev AP:54.5814, 46.2114, 40.4017
3d AP:47.3604, 36.0716, 31.0154
aos AP:65.17, 49.90, 48.77
Car AP_R40@0.70, 0.70, 0.70:
bbox AP:67.3332, 51.9216, 47.9794
bev AP:55.7742, 43.6306, 38.8716
3d AP:44.8888, 33.6227, 29.7601
aos AP:66.07, 50.69, 46.51
Car AP@0.70, 0.50, 0.50:
bbox AP:66.2887, 50.9406, 50.1121
bev AP:68.2305, 58.6677, 51.9742
3d AP:67.9217, 57.8087, 51.2324
aos AP:65.17, 49.90, 48.77
Car AP_R40@0.70, 0.50, 0.50:
bbox AP:67.3332, 51.9216, 47.9794
bev AP:70.9956, 55.7373, 51.5859
3d AP:70.3942, 55.2152, 50.9204
aos AP:66.07, 50.69, 46.51
Pedestrian AP@0.50, 0.50, 0.50:
bbox AP:33.6640, 29.2730, 27.3461
bev AP:27.9636, 24.0015, 22.7002
3d AP:24.5179, 21.3948, 20.0227
aos AP:20.91, 18.85, 17.75
Pedestrian AP_R40@0.50, 0.50, 0.50:
bbox AP:30.7358, 25.6456, 23.6849
bev AP:24.0143, 19.3956, 18.1442
3d AP:20.2534, 16.4522, 14.7456
aos AP:15.12, 12.55, 11.53
Pedestrian AP@0.50, 0.25, 0.25:
bbox AP:33.6640, 29.2730, 27.3461
bev AP:39.6583, 34.1208, 32.0018
3d AP:39.6332, 34.0367, 31.9339
aos AP:20.91, 18.85, 17.75
Pedestrian AP_R40@0.50, 0.25, 0.25:
bbox AP:30.7358, 25.6456, 23.6849
bev AP:37.2498, 31.2550, 28.8028
3d AP:37.2306, 31.1573, 28.7183
aos AP:15.12, 12.55, 11.53
Cyclist AP@0.50, 0.50, 0.50:
bbox AP:46.2147, 37.0968, 35.8150
bev AP:41.0650, 32.3202, 31.0648
3d AP:39.3470, 31.4047, 29.8415
aos AP:34.72, 27.52, 26.69
Cyclist AP_R40@0.50, 0.50, 0.50:
bbox AP:44.9303, 34.4198, 33.3809
bev AP:39.6938, 29.3634, 27.9016
3d AP:37.9765, 28.0044, 26.6262
aos AP:31.65, 22.93, 22.31
Cyclist AP@0.50, 0.25, 0.25:
bbox AP:46.2147, 37.0968, 35.8150
bev AP:47.3290, 37.4161, 35.6393
3d AP:47.2946, 37.4113, 35.6393
aos AP:34.72, 27.52, 26.69
Cyclist AP_R40@0.50, 0.25, 0.25:
bbox AP:44.9303, 34.4198, 33.3809
bev AP:46.0618, 34.6406, 33.3011
3d AP:45.9248, 34.6362, 33.2539
aos AP:31.65, 22.93, 22.31
I'm following the code and config file you provided on the KITTI data and it still doesn't seem to achieve the performance on the validation set you mentioned in your paper, can you provide me with your checkpoint or are there any other tuning parameters? And I found that the dynamic group configuration in the code repository is not used now, is it also not used in your best checkpoint. I hope you can answer this!
Hello, I am a beginner and I have a question for you, about submitting on the official website of kitti dataset to get the test set results, is there a result.txt file that needs to be submitted, and how do I get this file on my own server?
Hi,
Thank you for your work,
I'm currently trying to migrate your model to the latest openpcdet version. I have currently done the following:
extra_compile_args
: extra_compile_args = {
"cxx": ["-O3"],
"nvcc": [
"-O3",
"-DCUDA_HAS_FP16=1",
"-D__CUDA_NO_HALF_OPERATORS__",
"-D__CUDA_NO_HALF_CONVERSIONS__",
"-D__CUDA_NO_HALF2_OPERATORS__",
"-lcudnn",
]
}
and make_cuda_ext
the sparse operations:
make_cuda_ext(
name='pointnet2_batch_cuda',
module='pcdet.ops.pointnet2.pointnet2_batch',
sources=[
'src/pointnet2_api.cpp',
'src/ball_query.cpp',
'src/ball_query_gpu.cu',
'src/group_points.cpp',
'src/group_points_gpu.cu',
'src/interpolate.cpp',
'src/interpolate_gpu.cu',
'src/sampling.cpp',
'src/sampling_gpu.cu',
'src/ball_query_sparse.cpp',
'src/ball_query_sparse_gpu.cu',
'src/group_points_sparse.cpp',
'src/group_points_sparse_gpu.cu',
'src/sparse_indexing.cpp',
'src/sparse_indexing_gpu.cu',
],
),
since THC
has been deprecated in newer versions of pytorch, I have commented them out:
#include <torch/serialize/tensor.h>
#include <cuda.h>
#include <cuda_runtime_api.h>
#include <vector>
// #include <THC/THC.h>
#include "group_points_sparse_gpu.h"
// extern THCState *state;
// ...
running python3 setup.py develop
also doesn't give any problems, but when I try training, I get the following error:
/home/user/OpenPCDet/pcdet/ops/pointnet2/pointnet2_batch/pointnet2_utils.py", line 329, in forward
pointnet2.ball_query_sparse_wrapper(B, N, M, K, radius, nsample, new_xyz, xyz, indices, idx)
AttributeError: module 'pcdet.ops.pointnet2.pointnet2_batch.pointnet2_batch_cuda' has no attribute 'ball_query_sparse_wrapper'
Any ideas how I can fix this problem? Thanks
Thanks for great work!In the paper, it is mentioned that "we introduce the dynamic network mechanism into the IA-SSD framework". But in your repo, it seems that you introduce the dynamic network mechanism into the PointNet++.
Which one is used for the experiment results in your paper? Can you provide the code based on IA-SSD framework?
Hi, is the checkpoint of the model missing? I cant seem to find DBQ-SSD.pth
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.