Awesome-Point-Clouds

Awesome paper/code for point clouds with deep learning methods in detection and tracking. If you find some novel methods or have suggestions, please contact ybcui95@163.com

Datasets

• KITTI: Are we ready for Autonomous Driving? The KITTI Vision Benchmark Suite [Project Page] [CVPR'2012]
• Apolloscape: The ApolloScape Dataset for Autonomous Driving [Project Page] [CVPR'2018]
• Argoverse: Argoverse: 3D Tracking and Forecasting with Rich Maps [Project Page] [CVPR'2019]
• Nuscenes: nuScenes: A multimodal dataset for autonomous driving [Project Page] [arXiv'2019]
• H3D: The H3D Dataset for Full-Surround 3D Multi-Object Detection and Tracking in Crowded Urban Scenes [Project Page] [ICRA'2019]
• BLVD: BLVD: Building A Large-scale 5D Semantics Benchmark for Autonomous Driving [Project Page] [ICRA'2019]
• Waymo: Scalability in Perception for Autonomous Driving: Waymo Open Dataset [Project Page] [arXiv'2019]
• A* 3D: A* 3D: An Autonomous Driving Dataset in Challeging Environments [Project Page] [ICRA'2020]
• Ford AV Dataset : Ford Multi-AV Seasonal Dataset [Project Page] [arXiv'2020]
• A2D2 : A2D2: Audi Autonomous Driving Dataset [Project Page] [arXiv'2020]
• ONCE : One Million Scenes for Autonomous Driving: ONCE Dataset [Project Page] [NeurIPS'2021]
• Argoverse 2 : Argoverse 2: Next Generation Datasets for Self-Driving Perception and Forecasting [Project Page] [NeurIPS'2021]
• DAIR-V2X : DAIR-V2X : A Large-Scale Dataset for Vehicle-Infrastructure Cooperative 3D Object Detection [Project Page] [CVPR'2022]

Detection

Toolbox

• MMDetection3D: MMDetection3D is an open source object detection toolbox based on PyTorch, towards the next-generation platform for general 3D detection.
• OpenPCDet: OpenPCDet is a clear, simple, self-contained open source project for LiDAR-based 3D object detection.

Outdoors

• MV3D: Multi-View 3D Object Detection Network for Autonomous Driving [Code] [CVPR'2017]
• Frustum-Pointnets: Frustum PointNets for 3D Object Detection from RGB-D Data [Code] [CVPR'2018]
• PIXOR: PIXOR: Real-time 3D Object Detection from Point Clouds [Code] [CVPR'2018]
• IPOD: IPOD: Intensive Point-based Object Detector for Point Cloud [arXiv'2018]
• VoxelNet: VoxelNet: End-to-End Learning for Point Cloud Based 3D Object Detection [Code] [CVPR'2018]
• FaF: Fast and Furious: Real Time End-to-End 3D Detection, Tracking and Motion Forecasting with a Single Convolutional Net [CVPR'2018]
• Second: SECOND: Sparsely Embedded Convolutional Detection [Code] [Sensors'2018]
• AVOD: Joint 3D Proposal Generation and Object Detection from View Aggregation [Code] [IROS'2018]
• RoarNet: RoarNet: A Robust 3D Object Detection based on RegiOn Approximation Refinement [Code] [IV'2019]
• PointPillars: PointPillars: Fast Encoders for Object Detection from Point Clouds [Code] [CVPR'2019]
• PointRCNN: PointRCNN: 3D Object Proposal Generation and Detection from Point Cloud [Code] [CVPR'2019]
• Pseudo-LiDAR: Pseudo-LiDAR from Visual Depth Estimation: Bridging the Gap in 3D Object Detection for Autonomous Driving [Code] [CVPR'2019]
• LaserNet: LaserNet: An Efficient Probabilistic 3D Object Detector for Autonomous Driving [CVPR'2019]
• LaserNet++: Sensor Fusion for Joint 3D Object Detection and Semantic Segmentation [CVPR'2019 Workshop]
• Fast Point R-CNN: Fast Point R-CNN [ICCV'2019]
• STD: STD: Sparse-to-Dense 3D Object Detector for Point Cloud [Code] [ICCV'2019]
• PointPainting: PointPainting: Sequential Fusion for 3D Object Detection[arXiv'2019]
• Part-A^2: From Points to Parts: 3D Object Detection from Point Cloud with Part-aware and Part-aggregation Network [Code] [TPAMI'2020]
• Pseudo-LiDAR++: Pseudo-LiDAR++: Accurate Depth for 3D Object Detection in Autonomous Driving [Code] [ICLR'2020]
• PV-RCNN: PV-RCNN: Point-Voxel Feature Set Abstraction for 3D Object Detection [Code] [CVPR'2020]
• Point-GNN: Point-GNN: Graph Neural Network for 3D Object Detection in a Point Cloud [Code] [CVPR'2020]
• SA-SSD: Structure Aware Single-stage 3D Object Detection from Point Cloud [Code] [CVPR'2020]
• 3DSSD: 3DSSD: Point-based 3D Single Stage Object Detector [Code] [CVPR'2020]
• EPNet: EPNet: Enhancing Point Features with Image Semantics for 3D Object Detection [Code] [ECCV'2020]
• HotSpot: Object as Hotspots: An Anchor-Free 3D Object Detection Approach via Firing of Hotspots [ECCV'2020]
• Pillar-MVF: Pillar-based Object Detection for Autonomous Driving [ECCV'2020]
• Pyramid R-CNN: Pyramid R-CNN: Towards Better Performance and Adaptability for 3D Object Detection [Code] [ICCV'2021]
• VoTr: Voxel Transformer for 3D Object Detection [Code] [ICCV'2021]
• 4D-Net: 4D-Net for Learned Multi-Modal Alignment [ICCV'2021]
• RangeDet: RangeDet: In Defense of Range View for LiDAR-based 3D Object Detection [Code] [ICCV'2021]
• LiDAR R-CNN: LiDAR R-CNN: An Efficient and Universal 3D Object Detector [CVPR'2021]
• MVP: Multimodal Virtual Point 3D Detection [Code] [NeurIPS'2021]
• RSN: RSN: Range Sparse Net for Efficient, Accurate LiDAR 3D Object Detection [CVPR'2021]
• PDV: Point Density-Aware Voxels for LiDAR 3D Object Detection [Code] [CVPR'2022]
• VoxSeT: Voxel Set Transformer: A Set-to-Set Approach to 3D Object Detection from Point Clouds [Code] [CVPR'2022]
• SST: Embracing Single Stride 3D Object Detector with Sparse Transformer [Code] [CVPR'2022]
• TransFusion: TransFusion: Robust LiDAR-Camera Fusion for 3D Object Detection with Transformers [Code] [CVPR'2022]
• Point2Seq: Point2Seq: Detecting 3D Objects as Sequences [Code] [CVPR'2022]
• PillarNet: PillarNet: Real-Time and High-Performance Pillar-based 3D Object Detection [Code] [ECCV'2022]
• CenterFormer: CenterFormer: Center-based Transformer for 3D Object Detection [Code] [ECCV'2022]
• Pillar R-CNN: Pillar R-CNN for Point Cloud 3D Object Detection [Code] [arXiv'2023]
• DSVT: DSVT: Dynamic Sparse Voxel Transformer with Rotated Sets [Code] [CVPR'2023]
• FlatFormer: FlatFormer: Flattened Window Attention for Efficient Point Cloud Transformer [Code] [CVPR'2023]
• VoxelNeXt: VoxelNeXt: Fully Sparse VoxelNet for 3D Object Detection and Tracking [Code] [CVPR'2023]
• LinK: LinK: Linear Kernel for LiDAR-based 3D Perception [Code] [CVPR'2023]
• LargeKernel3D: LargeKernel3D:Scaling up Kernels in 3D CNNs [Code] [CVPR'2023]
• ConQueR: ConQueR: Query Contrast Voxel-DETR for 3D Object Detection [Code] [CVPR'2023]
• FocalFormer3D: FocalFormer3D: Focusing on Hard Instance for 3D Object Detection [Code] [ICCV'2023]
• Li3DeTr: Li3DeTr: A LiDAR based 3D Detection Transformer [WACV'2023]

Indoors

• VoteNet: Deep Hough Voting for 3D Object Detection in Point Clouds [Code] [ICCV'2019]
• ImVoteNet: ImVoteNet: Boosting 3D Object Detection in Point Clouds with Image Votes [Code] [CVPR'2020]
• 3DETR: An End-to-End Transformer Model for 3D Object Detection [Code] [ICCV'2021]
• VENet: VENet: Voting Enhancement Network for 3D Object Detection [ICCV'2021]
• Group-Free: Group-Free 3D Object Detection via Transformers [Code] [ICCV'2021]
• RBGNet: RBGNet: Ray-based Grouping for 3D Object Detection [Code] [CVPR'2022]
• FCAF3D: FCAF3D: Fully Convolutional Anchor-Free 3D Object Detection [Code] [ECCV'2022]

Tracking

MOT

• Complexer-YOLO: Complexer-YOLO: Real-Time 3D Object Detection and Tracking on Semantic Point Clouds [Code] [CVPR'2019 Workshop]
• 3DSiamese: Leveraging Shape Completion for 3D Siamese Tracking [Code] [CVPR'2019]
• AB3DMOT: A Baseline for 3D Multi-Object Tracking [Code] [arXiv'2019]
• mmMOT: Robust Multi-Modality Multi-Object Tracking [Code] [ICCV'2019]
• DSM: End-to-end Learning of Multi-sensor 3D Tracking by Detection [ICRA'2019]
• PointTrackNet: PointTrackNet: An End-to-End Network for 3-D Object Detection and Tracking from Point Clouds [ICRA'2020]
• Mahalanobis-KF: Probabilistic 3D Multi-Object Tracking for Autonomous Driving [Code] [arXiv'2020]
• GNN3DMOT: GNN3DMOT: Graph Neural Network for 3D Multi-Object Tracking with 2D-3D Multi-Feature Learning [Code] [CVPR'2020]
• PC-TCNN: Tracklet Proposal Network for Multi-Object Tracking on Point Clouds [IJCAI'2021]
• LOGR: Learnable Online Graph Representations for 3D Multi-Object Tracking [arXiv'2021]
• CenterPoint: Center-based 3D Object Detection and Tracking [Code] [CVPR'2021]
• Immortal-Tracker: Immortal Tracker: Tracklet Never Dies [Code] [arXiv'2021]
• SimpleTrack: SimpleTrack: Understanding and Rethinking 3D Multi-object Tracking [Code] [arXiv'2021]

SOT

• P2B: P2B: Point-to-Box Network for 3D Object Tracking in Point Clouds [Code] [CVPR'2020]
• PTT: PTT: Point-Track-Transformer Module for 3D Single Object Tracking in Point Clouds [Code] [IROS'2021]
• BAT: Box-Aware Feature Enhancement for Single Object Tracking on Point Clouds [Code] [ICCV'2021]
• LTTR: 3D Object Tracking with Transformer [Code] [BMVC'2021]
• V2B: 3D Siamese Voxel-to-BEV Tracker for Sparse Point Clouds [Code] [NeurIPS'2021]
• PTTR: PTTR: Relational 3D Point Cloud Object Tracking with Transformer [Code] [CVPR'2022]
• M2-Tracker: Beyond 3D Siamese Tracking: A Motion-Centric Paradigm for 3D Single Object Tracking in Point Clouds [Code] [CVPR'2022]
• STNet: 3D Siamese Transformer Network for Single Object Tracking on Point Clouds [Code] [ECCV'2022]
• SMAT: Exploiting More Information in Sparse Point Cloud for 3D Single Object Tracking [Code] [RAL]
• STTracker: STTracker: Spatio-Temporal Tracker for 3D Single Object Tracking [RAL]
• GLT-T: GLT-T: Global-Local Transformer Voting for 3D Single Object Tracking in Point Clouds [Code] [AAAI'2023]
• CXTrack: CXTrack: Improving 3D Point Cloud Tracking with Contextual Information [Code] [CVPR'2023]
• SyncTrack: Synchronize Feature Extracting and Matching: A Single Branch Framework for 3D Object Tracking [ICCV'2023]
• MBPTrack: MBPTrack: Improving 3D Point Cloud Tracking with Memory Networks and Box Priors [Code] [ICCV'2023]