Point Cloud and Sensor Fusion Materials

3D Point Cloud Ground Segmentation with code:

Collect paper about ground segmentation in 3D point cloud. This is the first and a crucial step towards object detection of 3d Point Clouds.

Geometry and ML Based With Code:

• Fast Segmentation of 3D Point Clouds for Ground Vehicles (2010) [paper], [code], [3rd party implementation]

• (GP-INASC) On the Segmentation of 3D LIDAR Point Clouds (2011) [paper] [code]

• Fast Segmentation of 3D Point Clouds: A Paradigm on LiDAR Data for Autonomous Vehicle Applications (2017) [paper] [single plane code], [multi plane code], [3rd party implementation]

• A Slope-robust Cascaded Ground Segmentation in 3D Point Cloud for Autonomous Vehicles (2018) [paper], [Python], [c++]

• A Probability Occupancy Grid Based Approach for Real-Time LiDAR Ground Segmentation (2019) [paper], [3rd party implementation]

• Patchwork: Concentric Zone-based Region-wise Ground Segmentation with Ground Likelihood Estimation Using a 3D LiDAR Sensor (2021) [paper], [code]

• (RECM-JCP) Fast Ground Segmentation for 3D LiDAR Point Cloud Based on Jump-Convolution-Process (2021) [paper], [code]

• Patchwork++: Fast and Robust Ground Segmentation Solving Partial Under-Segmentation Using 3D Point Cloud (2022) [paper], [code]

• GndNet: Fast Ground Plane Estimation and Point Cloud Segmentation for Autonomous Vehicles [code]

Elevation and terrain mapping for Robotics:

The elevation mapping technique which was introduced in 2005 during the DARPA challenge by [Researchers at Stanford University] was proposed for easier representation of point clouds for ground segementation and real-time autonomous driving. Although, some of the above indicated in literature the use of an elevation method [RECM-JCP], a review of thier code shows the use of the polar grid mapping method which is also a very popular and efficient technique for point cloud representation. This section would also include elevation and terrain mapping applied to SLAM and robotic navigation.

• Probabilistic Terrain Mapping for Mobile Robots With Uncertain Localization (2018) [paper], [code]

• GEM: Online Globally Consistent Dense Elevation Mapping for Unstructured Terrain [paper], [code]

• Elevation Mapping for Locomotion and Navigation using GPU [paper], [code]

• RING++: Roto-translation Invariant Gram for Global Localization on a Sparse Scan Map [ring] [ring++], [code]

• Reconstructing occluded Elevation Information in Terrain Maps with Self-supervised Learning (2022) [paper], [code]

• Terrain mapping algorithm for motion planning and control by [robot locomotion]

• [paper], [code]

Metrics

• Ground segmentation benchmark in SemanticKITTI dataset by [url-kaist team]

2. Clustering Algorithms and Techniques

• A Technical Survey and Evaluation of Traditional Point Cloud Clustering Methods for LiDAR Panoptic Segmentation (2021) [paper], [code]

• FEC: Fast Euclidean Clustering for Point Cloud Segmentation [paper], [code]

2. Deep Learning Segmentation and Object detection Algorithms

• PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation. [paper] [code]
• PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space. [paper] [code]
• Frustum PointNets for 3D Object Detection from RGB-D Data [paper] [code]
• RTM3D: Real-time Monocular 3D Detection from Object Keypoints for Autonomous Driving" (ECCV 2020) [paper] [code] [code]
• Complex-YOLO: An Euler-Region-Proposal for Real-time 3D Object Detection on Point Clouds [paper] [code]
• Super Fast and Accurate 3D Object Detection based on 3D LiDAR Point Clouds (The PyTorch implementation) [code]
• VoxelNet: End-to-End Learning for Point Cloud Based 3D Object Detection [paper] [code] [code] [code] [code] [code]
• Multi-View 3D Object Detection Network for Autonomous Driving [paper] [code]
• Lightweight and Accurate Point Cloud Clustering. [paper] [code]
• Linked Dynamic Graph CNN: Learning through Point Cloud by Linking Hierarchical Features. [paper] [code]
• Point-to-Voxel Knowledge Distillation for LiDAR Semantic Segmentation. [paper] [paper] [code]
• Point Transformer. [paper] [paper] [code] [code]
• GndNet: Fast Ground Plane Estimation and Point Cloud Segmentation for Autonomous Vehicles. [paper] [code]
• DeltaConv: Anisotropic Operators for Geometric Deep Learning on Point Clouds. [paper] [code]
• PointSIFT: A SIFT-like Network Module for 3D Point Cloud Semantic Segmentation. [paper] [code]
• RangeNet++: Fast and Accurate LiDAR Semantic Segmentation. [paper] [code]
• Learning Geometry-Disentangled Representation for Complementary Understanding of 3D Object Point Cloud. [paper] [code]
• Dynamic Graph CNN for Learning on Point Clouds. [paper [code]
• PointConv: Deep Convolutional Networks on 3D Point Clouds. [paper] [code]
• PointNetLK: Robust & Efficient Point Cloud Registration using PointNet. [paper] [code]
• PCN: Point Completion Network. [paper] [code]
• RPM-Net: Robust Point Matching using Learned Features. [paper] [code]
• 3D ShapeNets: A Deep Representation for Volumetric Shapes. [paper] [code]
• Correspondence Matrices are Underrated. [paper] [code]
• MaskNet: A Fully-Convolutional Network to Estimate Inlier Points. [paper] [code]
• 3DLineDetection. [paper] [code]
• Deep Learning for 3D Point Clouds: A Survey (IEEE TPAMI, 2020). [paper] [code]
• LatticeNet: Fast Point Cloud Segmentation Using Permutohedral Lattices. [paper] [code]
• SqueezeSegV3: Spatially-Adaptive Convolution for Efficient Point-Cloud Segmentation. [paper] [code]
• Monte Carlo Convolution for Learning on Non-Uniformly Sampled Point Clouds. [paper] [code]
• cilantro: A Lean, Versatile, and Efficient Library for Point Cloud Data Processing.
• Oriented Point Sampling for Plane Detection in Unorganized Point Clouds. [paper] [code]
• Supervoxel for 3D point clouds. [paper] [code]
• MmWave Radar Point Cloud Segmentation using GMM in Multimodal Traffic Monitoring. [paper] [code]
• LRGNet: Learnable Region Growing for Class-Agnostic Point Cloud Segmentation. [paper] [code]
• PAConv: Position Adaptive Convolution with Dynamic Kernel Assembling on Point Clouds. [paper] [code]
• SCF-Net: Learning Spatial Contextual Features for Large-Scale Point Cloud Segmentation. [paper] [code]
• Point-BERT: Pre-training 3D Point Cloud Transformers with Masked Point Modeling. [paper] [code]
• RETHINKING NETWORK DESIGN AND LOCAL GEOMETRY IN POINT CLOUD: A SIMPLE RESIDUAL MLP FRAMEWORK. [paper] [code]
• Masked AutoenGitHubrs for Point Cloud Self-supervised Learning. [paper] [code]
• EagerMOT: 3D Multi-Object Tracking via Sensor Fusion [paper] [code]
• PointPainting: Sequential Fusion for 3D Object Detection [paper] [code] [code]
• Automatic Radar-Camera Dataset Generation for Sensor-Fusion Applications [paper] [code]
• Radar Voxel Fusion for 3D Object Detection [paper] [code]
• CRF-Net for Object Detection (Camera and Radar Fusion Network) [paper] [code]
• [paper] [code]

Others

• LCCNet: LiDAR and Camera Self-Calibration using Cost Volume Network [paper] [code]
• A Collection of LiDAR-Camera-Calibration Papers, Toolboxes and Notes [code]
• LiDAR-camera system extrinsic calibration by establishing virtual point correspondences from pseudo calibration objects [paper]
• A list of papers and datasets about point cloud analysis (processing) [code]
• ICCV-2021-point-cloud-analysis [code]
• Lidar and radar fusion for real-time road-objects detection and tracking [paper]
• Awesome Radar Perception [code]
• Automatic Extrinsic Calibration of Vision and Lidar by Maximizing Mutual Information [paper]
• Surrounding Objects Detection and Tracking for Autonomous Driving Using LiDAR and Radar Fusion [paper]
• Camera-LiDAR Multi-Level Sensor Fusion for Target Detection at the Network Edge [paper]
• Real-time RADAR and LIDAR Sensor Fusion for Automated Driving [paper]
• Radar Camera Fusion via Representation Learning in Autonomous Driving [paper]

Perception Resource

• Pytorch C++ by [prabhuomkar]
• Paper with Code [site]
• 3D point cloud by [zhulf0804]
• Lidar-Ground-Segmantation-Paper-List by [wangx1996]
• The list of vision-based SLAM by [tzutalin]
• Lidar Point clound processing for Autonomous Driving by [beedotkiran]
• Vision-Centric-BEV-Perception by [4DVLab]
• Awesome BEV Perception from Multi-Cameras by [chaytonmin]
• Awesome Radar Perception by [ZHOUYI1023]

Useful Links

• IoT: Awesome edge computing by [qijianpeng]
• Logical Neural Networks a neuro symbolic framework by [IBM]
• DeepSpeed Examples by [microsoft]
• GPT4All: An ecosystem of open-source on-edge large language models by [nomic-ai]