Skip to content

Latest commit

 

History

History
132 lines (114 loc) · 13.8 KB

3D-Object-Detection.md

File metadata and controls

132 lines (114 loc) · 13.8 KB
Raw

3D Object Detection: An Overview

This repository provides an up-to-date list of 3D Object Detection works.

=============================

Table of contents

  1. Lidar based 3D Object Detection
    1.1 Point Cloud based
    1.2 Voxel based
    1.3 Range Image based
    1.4 Detection and Tracking
    1.5 Graph based
    1.6 Ground Plane Detection
    1.7 Fast Object Detection
    1.8.Evaluation Metric
  2. Camera based 3D Object Detection
    2.1 xxxxxx
    2.2 xxxxxx
    2.3 xxxxx
  3. Fusion based 3D Object Detection
    3.1 xxxxx
    3.2 xxxxx
    3.3 xxxxx

1. Lidar based 3D Object Detection

1.1 Point Cloud based

  • SE-SSD: Self-Ensembling Single-Stage Object Detector From Point Cloud; [Paper] [Code]
  • Voxel R-CNN:Towards High Performance Voxel-based 3D Object Detection; [Paper] [Code]
  • 3DSSD: Point-based 3D Single Stage Object Detector; [Paper] [Code]
  • SASA: Semantics-Augmented Set Abstraction for Point-based 3D Object Detection; [Paper] [Code]
  • Not All Points Are Equal: Learning Highly Efficient Point-based Detectors for 3D LiDAR Point Clouds; [Paper] [Code]
  • Focal Sparse Convolutional Networks for 3D Object Detection; [Paper] [Code]
  • LiDAR Distillation: Bridging the Beam-Induced Domain Gap for 3D Object Detection; [Paper] [Code]
  • Point Density-Aware Voxels for LiDAR 3D Object Detection; [Paper] [Code]
  • OccAM’s Laser: Occlusion-based Attribution Maps for 3D Object Detectors on LiDAR Data; [Paper] [Code]
  • Pillar-based Object Detection for Autonomous Driving; [Paper] [Code]
  • Behind the Curtain: Learning Occluded Shapes for 3D Object Detection; [Paper] [Code]
  • An LSTM Approach to Temporal3D Object Detection in LiDAR P oint Clouds; [Paper] [Code]
  • A VERSATILE MULTI-VIEW FRAMEWORK FOR LIDAR-BASED 3D OBJECT DETECTION WITH GUIDANCE FROM PANOPTIC SEGMENTATION [Paper]
  • Point2Seq: Detecting 3D Objects as Sequences; [Paper]
  • Super Fast and Accurate 3D Object Detection based on 3D LiDAR Point Clouds; [Code]
  • Identifying Unknown Instances for Autonomous Driving; [Paper]
  • Deep Multi-Sensor Lane Detection; [Paper]
  • Fast and Furious: Real Time End-to-End 3D Detection, Tracking and Motion Forecasting with a Single Convolutional Net; [Paper]
  • PIXOR: Real-time 3D Object Detection from Point Clouds; [Paper]
  • Efficient Convolutions for Real-Time Semantic Segmentation of 3D Point Clouds; [Paper]
  • Open-set 3D Object Detection; [Paper]
  • ORDER: Open World Object Detectionon Road Scenes; [Paper]
  • Towards Open-Set Object Detection and Discovery; [Paper]
  • 2022 RBGNet: Ray-based Grouping for 3D Object Detection; [Paper] [Code]
  • 2019 Deep Hough Voting for 3D Object Detection in Point Clouds; [Paper] [Code]
  • 2019 Fast Point R-CNN; [Paper]

1.2 Voxel based

  • 2022 A Unified Query-based Paradigm for Point Cloud Understanding; [Paper]
  • 2022 PillarNet: Real-Time and High-Performance Pillar-based 3D Object Detection; [Paper]
  • 2021 PV-RCNN++: Point-Voxel Feature Set Abstraction With Local Vector Representation for 3D Object Detection; [Paper] [Code]
  • 2021 BEVDetNet: Bird’s Eye View LiDAR Point Cloud based Real-time 3D Object Detection for Autonomous Driving; [Paper]
  • 2021 DV-Det:Efficient 3D Point Cloud Object Detection with Dynamic Voxelization [Paper]
  • 2021 Anchor-free 3D Single Stage Detector with Mask-Guided Attention for Point Cloud; [Paper] [Code]
  • 2021 Improved Pillar with Fine-grained Feature for 3D Object Detection; [Paper]
  • 2021 FCAF3D: Fully Convolutional Anchor-Free 3D Object Detection; [Paper] [Code]
  • 2021 PiFeNet: Pillar-Feature Network for Real-Time 3D Pedestrian Detection from Point Cloud; [Paper]
  • 2021 Point Density-Aware Voxels for LiDAR 3D Object Detection; [Paper]
  • 2019 STD: Sparse-to-Dense 3D Object Detector for Point Cloud; [Paper]
  • 2021 PV-RCNN: Point-Voxel Feature Set Abstraction for 3D Object Detection; [Paper]
  • 2021 Center-based 3D Object Detection and Tracking; [Paper] [Code]
  • 2020 AFDet: Anchor Free One Stage 3D Object Detection; [Paper]
  • 2022 AFDetV2: Rethinking the Necessity of the Second Stage for Object Detection from Point Clouds; [Paper]
  • 2020 Dynamic Edge Weights in Graph Neural Networks for 3D Object Detection; [Paper]
  • 2020 CIA-SSD: Confident IoU-Aware Single-Stage Object Detector From Point Cloud; [Paper] [Code]

1.3 Range Image based

  • RangeSeg: Range-Aware Real Time Segmentation of 3D LiDAR Point Clouds; [Paper]
  • To the Point: Efficient 3D Object Detection in the Range Image with Graph Convolution Kernels; [Paper]
  • RSN: Range Sparse Net for Efficient, Accurate LiDAR 3D Object Detection; [Paper]
  • RangeDet: In Defense of Range View for LiDAR-based 3D Object Detection; [Paper]
  • RangeRCNN: Towards Fast and Accurate 3D Object Detection with Range Image Representation; [Paper]
  • Range Conditioned Dilated Convolutions for Scale Invariant 3D Object Detection; [Paper]

1.4 Detection and Tracking

  • Exploring Simple 3D Multi-Object Tracking for Autonomous Driving; [Paper] [Code]
  • You Don't Only Look Once: Constructing Spatial-Temporal Memory for Integrated 3D Object Detection and Tracking; [Paper] [Code]
  • Beyond 3D Siamese Tracking: A Motion-Centric Paradigm for 3D Single Object Tracking in Point Clouds; [Paper] [Code]
  • TPCN: Temporal Point Cloud Networks for Motion Forecasting; [Paper]
  • PointTrackNet: An End-to-End Network for 3-D Object Detection and Tracking from Point Clouds; [Paper]
  • AB3DMOT: A Baseline for 3D Multi-Object Tracking and New Evaluation Metrics; [Paper] [Code]

1.5 Graph based

  • Grid-GCN for Fast and Scalable Point Cloud Learning; [Paper] [Code]
  • Point-GNN: Graph Neural Network for 3D Object Detection in a Point Cloud; [Paper] [Code]
  • Dynamic Graph CNN for Learning on Point Clouds; [Paper] [Code]
  • 2021 Probabilistic 3D Multi-Modal, Multi-Object Tracking for Autonomous Driving; [Paper]

1.6 Ground Plane Detection

  • GndNet: Fast Ground Plane Estimation and Point Cloud Segmentation for Autonomous Vehicles; [Paper] [Code]
    • How to Build a Curb Dataset with LiDAR Data for Autonomous Driving; [Paper] [Dataset]
    • CurbScan: Curb Detection and Tracking Using Multi-Sensor Fusion; [Paper] [Dataset]
    • Lidar-histogram for fast road and obstacle detection; [Paper]

1.7 Fast Object Detection

  • 2022 AFDetV2: Rethinking the Necessity of the Second Stage for Object Detection from Point Clouds; [Paper]
  • 2022 PillarNet: Real-Time and High-Performance Pillar-based 3D Object Detection; [Paper]
  • 2021 Real-Time Anchor-Free Single-Stage 3D Detection with IoU-Awareness; [Paper]
  • 2020 AFDet: Anchor Free One Stage 3D Object Detection; [Paper]
  • 2020 CIA-SSD: Confident IoU-Aware Single-Stage Object Detector From Point Cloud; [Paper] [Code]
  • 2021 1 st Place Solution for Waymo Open Dataset Challenge - 3D Detection and Domain Adaptation; [Paper]
  • 2021 CenterPoint: Center-based 3D Object Detection and Tracking; [Paper] [Code] [TensorRT implementation] [ONNX and TensorRT]
  • 2021 CenterPoint++ submission to the Waymo Real-time 3D Detection Challenge; [Paper] [Code]
  • 2021 RSN: Range Sparse Net for Efficient, Accurate LiDAR 3D Object Detection; [Paper]
  • 2021 HVPR: Hybrid Voxel-Point Representation for Single-stage 3D Object Detection; [Paper]
  • 2021 SRDAN: Scale-aware and Range-aware Domain Adaptation Network for Cross-dataset 3D Object Detection; [Paper]
  • 2021 PointAugmenting: Cross-Modal Augmentation for 3D Object Detection; [Paper]
  • 2021 BEVDetNet: Bird’s Eye View LiDAR Point Cloud based Real-time 3D Object Detection for Autonomous Driving; [Paper]
  • 2021 RangeDet: In Defense of Range View for LiDAR-based 3D Object Detection; [Paper]
  • End-to-End Multi-View Fusion for 3D Object Detection in LiDAR Point Clouds; [Paper]

1.8 Evaluation Metric

  • 2020 Learning to Evaluate Perception Models Using Planner-Centric Metrics; [Paper]
  • 2021 The efficacy of Neural Planning Metrics: A meta-analysis of PKL on nuScenes; [Paper]
  • 2021 Revisiting 3D Object Detection From an Egocentric Perspective; [Paper]
  • 2020 nuScenes: A multimodal dataset for autonomous driving; [Paper]

https://github.com/beedotkiran/Lidar_For_AD_references