xinge008/Cylinder3D
Rank 1st in the leaderboard of SemanticKITTI semantic segmentation (both single-scan and multi-scan) (Nov. 2020) (CVPR2021 Oral)
PythonApache-2.0
Issues
- 1
About DDP
#172 opened by Someone-sleeping - 5
Bad Result at demo_folder.py
#152 opened by DonghoonPark12 - 1
How to see the demo effect?
#177 opened by Jeffgit2022 - 6
TypeError: No matching definition for argument type(s) array(uint8, 3d, C), array(int32, 2d, C)
#162 opened by Avril-Dragon - 3
validation Iou
#134 opened by fengjiang5 - 2
Details request
#171 opened by bhaba-ranjan - 0
about pretrained model
#176 opened by saligia-ira - 0
C++ version?
#175 opened by hilnkm - 0
Generalization Experiment-Related Questions
#174 opened by weixinhum2023 - 0
How to submit nuscenes results on Test set
#173 opened by Konstantin5389 - 3
- 2
problem of "ValueError: Expected input batch_size (2) to match target batch_size (1)."
#128 opened by zhaodejian - 5
About the weights in WCE
#169 opened by xpzbph - 0
Error when trying to re-train
#170 opened by SM1991CODES - 4
Mismatch in shape when running the demo
#149 opened by WEIJIARONG99 - 2
[Help wanted!] pretrained model mismatch
#137 opened by l9761116 - 2
not able to install prerequisite spconv v1.2.1
#148 opened by tarmas99 - 0
About multi-scan task on SemanticKITTI
#168 opened by liamlin5566 - 2
RuntimeError: radix_sort: failed on 1st step: cudaErrorUnsupportedPtxVersion: the provided PTX was compiled with an unsupported toolchain.
#164 opened by C-hongfei - 0
The size of cubic voxel in the paper
#167 opened by kimbss470 - 0
- 0
请问一下kitti的背景类为什么不需要训练
#165 opened by wfs123456 - 3
shortcut.features = self.act1(shortcut.features) ValueError: you can't set feature directly, use 'x = x.replace_feature(your_new_feature)' to generate new SparseConvTensor instead.
#135 opened by l9761116 - 0
How to output a shape of (N,1) ?
#163 opened by eyegotthis - 0
- 0
How to load a pretrained model?
#160 opened by rvzaytsev - 2
The link to the new Cylinder3D++ code is broken
#150 opened by chucktang98 - 1
Checkpoint with spconv2 for nuscenes dataset
#155 opened by chucktang98 - 0
- 0
怎么制作自己的数据集,我的数据是无人机激光雷达扫描的
#157 opened by yangfeifan-ufo - 0
No matching definition for argument type(s) array(uint8, 3d, C), array(int32, 2d, C)
#156 opened by yangfeifan-ufo - 1
- 2
- 1
Pointwise Cylinder Refinement Model
#151 opened by L-Reichardt - 1
docker build fail at step 33
#153 opened by thomas-enxuli - 0
data normalization
#147 opened by chky1997 - 0
A lightweight Cylinder3D model with much higher performance is now available!!!
#146 opened by cardwing - 0
circular_pad
#144 opened by YuQian19 - 3
Stuck when start training
#143 opened by mc171819 - 0
Code for detection
#142 opened by mc171819 - 0
In training loss
#141 opened by akshay-antony - 0
polar2cat function have a bug?
#140 opened by chuifengsimida - 1
Regarding other sparse conv libraries
#138 opened by akshay-antony - 1
Thanks for your great work. But it seems that there are many mismatches in the code.
#136 opened by l9761116 - 1
spconv version of Dockerfile
#133 opened by rdkouhei - 2
Cartesian and cylinder
#127 opened by RuixiangXue - 1
Why Asymmetrical block better than 3D CNN
#132 opened by Chuanyu-LD - 1
About visualization of voxelization
#129 opened by artofstate - 1
Cant find pretrained model for nuScenes
#131 opened by Serenos - 0