Robust 3D Semantic Segmentation

17 papers with code • 3 benchmarks • 3 datasets

3D Semantic Segmentation under Out-of-Distribution Scenarios


Use these libraries to find Robust 3D Semantic Segmentation models and implementations
4 papers
2 papers

Most implemented papers

KPConv: Flexible and Deformable Convolution for Point Clouds

HuguesTHOMAS/KPConv ICCV 2019

Furthermore, these locations are continuous in space and can be learned by the network.

4D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networks

StanfordVL/MinkowskiEngine CVPR 2019

To overcome challenges in the 4D space, we propose the hybrid kernel, a special case of the generalized sparse convolution, and the trilateral-stationary conditional random field that enforces spatio-temporal consistency in the 7D space-time-chroma space.

SqueezeSeg: Convolutional Neural Nets with Recurrent CRF for Real-Time Road-Object Segmentation from 3D LiDAR Point Cloud

BichenWuUCB/SqueezeSeg 19 Oct 2017

In this paper, we address semantic segmentation of road-objects from 3D LiDAR point clouds.

Searching Efficient 3D Architectures with Sparse Point-Voxel Convolution

mit-han-lab/spvnas ECCV 2020

Self-driving cars need to understand 3D scenes efficiently and accurately in order to drive safely.

SalsaNext: Fast, Uncertainty-aware Semantic Segmentation of LiDAR Point Clouds for Autonomous Driving

TiagoCortinhal/SalsaNext 7 Mar 2020

In this paper, we introduce SalsaNext for the uncertainty-aware semantic segmentation of a full 3D LiDAR point cloud in real-time.

PolarNet: An Improved Grid Representation for Online LiDAR Point Clouds Semantic Segmentation

edwardzhou130/PolarSeg CVPR 2020

The need for fine-grained perception in autonomous driving systems has resulted in recently increased research on online semantic segmentation of single-scan LiDAR.

CPGNet: Cascade Point-Grid Fusion Network for Real-Time LiDAR Semantic Segmentation

GangZhang842/CPGNet 21 Apr 2022

LiDAR semantic segmentation essential for advanced autonomous driving is required to be accurate, fast, and easy-deployed on mobile platforms.

Cylindrical and Asymmetrical 3D Convolution Networks for LiDAR Segmentation

xinge008/Cylinder3D CVPR 2021

However, we found that in the outdoor point cloud, the improvement obtained in this way is quite limited.

CENet: Toward Concise and Efficient LiDAR Semantic Segmentation for Autonomous Driving

huixiancheng/cenet 26 Jul 2022

Accurate and fast scene understanding is one of the challenging task for autonomous driving, which requires to take full advantage of LiDAR point clouds for semantic segmentation.