Robust 3D Semantic Segmentation On Nuscenes C

Metrics

mean Corruption Error (mCE)

Results

Performance results of various models on this benchmark

		Paper Title
FIDNet	122.42%	FIDNet: LiDAR Point Cloud Semantic Segmentation with Fully Interpolation Decoding
PolarNet	115.09%	PolarNet: An Improved Grid Representation for Online LiDAR Point Clouds Semantic Segmentation
CENet	112.79%	CENet: Toward Concise and Efficient LiDAR Semantic Segmentation for Autonomous Driving
Cylinder3D (spconv)	111.84%	Cylindrical and Asymmetrical 3D Convolution Networks for LiDAR Segmentation
WaffleIron	106.73%	Using a Waffle Iron for Automotive Point Cloud Semantic Segmentation
SPVCNN-18	106.65%	Searching Efficient 3D Architectures with Sparse Point-Voxel Convolution
Cylinder3D (torchsparse)	105.56%	Cylindrical and Asymmetrical 3D Convolution Networks for LiDAR Segmentation
MinkUNet-18	100.00%	4D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networks
2DPASS	98.56%	2DPASS: 2D Priors Assisted Semantic Segmentation on LiDAR Point Clouds
SPVCNN-34	97.45%	Searching Efficient 3D Architectures with Sparse Point-Voxel Convolution
MinkUNet-34	96.37%	4D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networks
GFNet	92.55%	GFNet: Geometric Flow Network for 3D Point Cloud Semantic Segmentation

0 of 12 row(s) selected.