Skip to content

This repository provides awesome research papers for autonomous driving perception. If you do find a problem or have any suggestions, please raise this as an issue or make a pull request with information (format of the repo): Research paper title, datasets, metrics, objects, source code, publisher, and year.

Notifications You must be signed in to change notification settings

maudzung/Awesome-Autonomous-Driving-Papers

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

1 Commit
 
 

Repository files navigation

Awesome-Autonomous-Driving-Papers


This repository provides awesome research papers for autonomous driving perception.
I have tried my best to keep this repository up to date. If you do find a problem or have any suggestions, please raise this as an issue or make a pull request with information (format of the repo): Research paper title, datasets, metrics, objects, source code, publisher, and year.

This summary is categorized into:

Abbreviations

  • AP-2D: Average Precision for 2D detection (on RGB-image space)
  • AP-3D: Average Precision for 3D detection
  • AP-BEV: Average Precision for Birds Eye View
  • AOS: Average Orientation Similarity (if 2D bounding box available)

Datasets

1. LiDAR-based 3D Object Detection

1.1 Single-stage detectors

Research Paper Datasets Metrics Objects Source Code Publisher Year
“HorizonLiDAR3D”: 1st Place Solution for Waymo Open Dataset Challenge - 3D Detection and Domain Adaptation
  • Waymo
  • AP-3D
  • AP-BEV
Cars, Pedestrians, Cyclists --- ArXiv 2020
Structure Aware Single-stage 3D Object Detection from Point Cloud (SA-SSD)
  • KITTI
  • AP-3D
  • AP-BEV
Cars PyTorch CVPR 2020
3DSSD: Point-based 3D Single Stage Object Detector
  • KITTI
  • nuScenes
  • AP-3D
  • AP-BEV
Cars PyTorch CVPR 2020
LaserNet: An Efficient Probabilistic 3D Object Detector for Autonomous Driving
  • KITTI
  • ATG4D
  • AP-3D
  • AP-BEV
Cars --- CVPR 2019
PointPillars: Fast Encoders for Object Detection from Point Clouds
  • KITTI
  • AP-3D
  • AP-BEV
Cars, Pedestrians, Cyclists PyTorch CVPR 2019
SECOND: Sparsely Embedded Convolutional Detection
  • KITTI
  • AP-3D
  • AP-BEV
Cars, Pedestrians, Cyclists PyTorch Sensors 2018
Complex-YOLO: an euler-region-proposal for real-time 3d object detection on point clouds
  • KITTI
  • AP-3D
  • AP-BEV
Cars, Pedestrians, Cyclists PyTorch ECCV 2018
YOLO3D: End-to-end real-time 3D Oriented Object Bounding Box Detection from LiDAR Point Cloud
  • KITTI
  • mAP-3D
Cars, Pedestrians, Cyclists PyTorch ECCV 2018
Pixor: Real-time 3d object detection from point clouds
  • KITTI
  • AP-3D
Cars PyTorch CVPR 2018
VoxelNet: End-to-End Learning for Point Cloud Based 3D Object Detection
  • KITTI
  • AP-3D
  • AP-BEV
Cars, Pedestrians, Cyclists PyTorch Tensorflow CVPR 2018
3D Fully Convolutional Network using PointCloud data for Vehicle Detection
  • KITTI
  • AP-3D
  • AOS
Cars Tensorflow IROS 2017

1.2 Two-stage detectors

Research Paper Datasets Metrics Objects Source Code Publisher Year
PV-RCNN: Point-Voxel Feature Set Abstraction for 3D Object Detection
  • KITTI
  • Waymo
  • AP-3D
Cars PyTorch CVPR 2020
Point-GNN: Graph Neural Network for 3D Object Detection in a Point Cloud
  • KITTI
  • AP-3D
  • AP-BEV
Cars (1 model), Pedestrians and Cyclists(1 model) Tensorflow CVPR 2020
3D IoU-Net: IoU Guided 3D Object Detector for Point Clouds
  • KITTI
  • AP-3D
  • AP-BEV
Cars --- ArXiv 2020
PointRCNN: 3D Object Proposal Generation and Detection from Point Cloud
  • KITTI
  • AP-3D
Cars, Pedestrians, Cyclists PyTorch CVPR 2019
Patch Refinement - Localized 3D Object Detection
  • KITTI
  • AP-3D
  • AP-BEV
Cars, Pedestrians, Cyclists --- ArXiv 2019
StarNet: Targeted Computation for Object Detection in Point Clouds
  • KITTI
  • Waymo
  • AP-3D
Cars, Pedestrians, Cyclists PyTorch ArXiv 2019
Part-A^2 Net: 3D Part-Aware and Aggregation Neural Network for Object Detection from Point Cloud
  • KITTI
  • AP-3D
  • AP-BEV
Cars, Pedestrians, Cyclists PyTorch ArXiv 2019
Fast Point R-CNN
  • KITTI
  • AP-3D
  • AP-BEV
Cars, Pedestrians, Cyclists --- ICCV 2019
STD: Sparse-to-Dense 3D Object Detector for Point Cloud
  • KITTI
  • AP-3D
Cars, Pedestrians, Cyclists --- ICCV 2019
Three-dimensional Backbone Network for 3D Object Detection in Traffic Scenes
  • KITTI
  • AP-3D
Cars, Pedestrians, Cyclists PyTorch ICCV 2019
Frustum ConvNet: Sliding Frustums to Aggregate Local Point-Wise Features for Amodal 3D Object Detection
  • KITTI
  • SUN-RGBD
  • AP-3D
  • AP-BEV
Cars, Pedestrians, Cyclists PyTorch IROS 2019
Frustum PointNets for 3D Object Detection from RGB-D Data
  • KITTI
  • SUN-RGBD
  • AP-3D
  • AP-BEV
Cars, Pedestrians, Cyclists Tensorflow CVPR 2018

2. Monocular Image-based 3D Object Detection

Research Paper Datasets Metrics Objects Source Code Publisher Year
RTM3D: Real-time Monocular 3D Detection from Object Keypoints for Autonomous Driving
  • KITTI
  • AP-3D
  • AP-BEV
  • AOS
Cars PyTorch ECCV 2020
Stereo R-CNN based 3D Object Detection for Autonomous Driving
  • KITTI
  • AP-3D/BEV
  • AP-2D
Cars PyTorch CVPR 2019
M3D-RPN: Monocular 3D Region Proposal Network for Object Detection
  • KITTI
  • AP-3D/BEV
  • AP-2D
Cars, Pedestrians, Cyclists PyTorch ICCV 2019
Mono3D++: Monocular 3D Vehicle Detection with Two-Scale 3D Hypotheses and Task Priors
  • KITTI
  • AP-3D/BEV
  • AP-2D
Cars, Pedestrians, Cyclists --- ArXiv 2019
3D Bounding Box Estimation Using Deep Learning and Geometry
  • KITTI
  • AP
Cars, Cyclists PyTorch CVPR 2017
Deep MANTA: A Coarse-to-fine Many-Task Network for joint 2D and 3D vehicle analysis from monocular image
  • KITTI
Cars --- CVPR 2017
Deep MANTA: A Coarse-to-fine Many-Task Network for joint 2D and 3D vehicle analysis from monocular image
  • KITTI
  • AP-2D/3D
  • AOS
Cars Link ICRA 2017

3. LiDAR and RGB Images Fusion

Research Paper Datasets Metrics Objects Source Code Publisher Year
ImVoteNet: Boosting 3D Object Detection in Point Clouds with Image Votes
  • SUN RGB-D
  • AP-3D
37 object categories PyTorch CVPR 2020
Multi-Task Multi-Sensor Fusion for 3D Object Detection
  • KITTI
  • AP-2D
  • AP-3D
  • AP-BEV
Cars, Pedestrians, Cyclists PyTorch CVPR 2019

4. Pseudo-LiDAR

Research Paper Datasets Metrics Objects Source Code Publisher Year
Pseudo-LiDAR++: Accurate Depth for 3D Object Detection in Autonomous Driving
  • KITTI
  • AP-3D
  • AP-BEV
Cars, Pedestrians, Cyclists PyTorch ICLR 2020
Pseudo-LiDAR from Visual Depth Estimation: Bridging the Gap in 3D Object Detection for Autonomous Driving
  • KITTI
  • AP-3D
  • AP-BEV
Cars, Pedestrians, Cyclists PyTorch CVPR 2019

5. Training tricks

Research Paper Datasets Metrics Objects Source Code Publisher Year
PPBA: Improving 3D Object Detection through Progressive Population Based Augmentation
  • KITTI
  • AP-3D
  • AP-BEV
Cars, Pedestrians, Cyclists --- ArXiv 2020
Class-balanced Grouping and Sampling for Point Cloud 3D Object Detection
  • KITTI
  • AP-3D
  • AP-BEV
10 object categories PyTorch ArXiv 2019
Weighted Point Cloud Augmentation for Neural Network Training Data Class-Imbalance
  • ScanNet
  • Semantic3D
    --- ArXiv 2019

    6. Object tracking (in progress)

    To do list:

    • Add 3D object detection papers based on LiDAR/monocular images/fusion/pseudo-LiDAR.
    • Add training tricks papers
    • Add object tracking papers.
    • Provide download.py script to automatically download .pdf files.

    References

    About

    This repository provides awesome research papers for autonomous driving perception. If you do find a problem or have any suggestions, please raise this as an issue or make a pull request with information (format of the repo): Research paper title, datasets, metrics, objects, source code, publisher, and year.

    Topics

    Resources

    Stars

    Watchers

    Forks