Awesome-Autonomous-Driving

Self-Driving Cars: A Survey [ ]

MIT Autonomous Vehicle Technology Study: Large-Scale Deep Learning Based Analysis of Driver Behavior and Interaction with Automation [ ]

github

Reading Material Advanced Machine Learning by SungjuHwang [ ]

Densely Connected Convolutional Networks [ ]

Microsoft (Deep Residual Learning) [ ][ ]

Microsoft (PReLu/Weight Initialization)

Batch Normalization

Differentiable Learning-to-Normalize via Switchable Normalization

GoogLeNet

VGG-Net

AlexNet

PVANET

OverFeat, NYU

R-CNN, UC Berkeley

SPP, Microsoft Research

Fast R-CNN, Microsoft Research

Faster R-CNN, Microsoft Research

R-CNN minus R, Oxford

End-to-end people detection in crowded scenes

You Only Look Once: Unified, Real-Time Object Detection , , ,

Inside-Outside Net

Deep Residual Network (Current State-of-the-Art)

Weakly Supervised Object Localization with Multi-fold Multiple Instance Learning [ ]

R-FCN

SSD

Speed/accuracy trade-offs for modern convolutional object detectors

SA General Pipeline for 3D Detection of Vehicles

Multi-Task Vehicle Detection With Region-of-Interest Voting

Car Detection for Autonomous Vehicle: LIDAR and Vision Fusion Approach Through Deep Learning Framework

A Unified Multi-scale Deep Convolutional Neural Network for Fast Object Detection

Complex-YOLO: An Euler-Region-Proposal for Real-time 3D Object Detection on Point Clouds

PIXOR: Real-time 3D Object Detection from Point Clouds

Fast and Furious: Real Time End-to-End 3D Detection, Tracking and Motion Forecasting with a Single Convolutional Net

Joint 3D Proposal Generation and Object Detection from ViewAggregation

Frustum PointNets for 3D Object Detection from RGB-D Data

PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation

PointNet++: Deep Hierarchical Feature Learning onPoint Sets in a Metric Space

Deep MANTA: A Coarse-to-fine Many-Task Network for joint 2D and 3D vehicleanalysis from monocular image

Monocular 3D Object Detection for Autonomous Driving

Subcategory-Aware Convolutional Neural Networks for Object Proposals and Detection

Beyond Pixels Leveraging Geometry and Shape Cues for Online Multi-Object Tracking

Tracking the Untrackable Learning to Track Multiple Cues with Long-Term Dependencies

Near-Online Multi-target Tracking with Aggregated Local Flow Descriptor,

Fast and Furious: Real Time End-to-End 3D Detection, Tracking and Motion Forecasting with a Single Convolutional Net

Liang-Chieh Chen+, George Papandreou+, Iasonas Kokkinos, Kevin Murphy, Alan L. Yuille (+ equal contribution). . In ICLR, 2015

Liang-Chieh Chen, George Papandreou, Florian Schroff, Hartwig Adam. . arXiv: 1706.05587, 2017

Liang-Chieh Chen, Yukun Zhu, George Papandreou, Florian Schroff, Hartwig Adam. arXiv: 1802.02611. . arXiv: 1802.02611, 2018

Wei Liu, Andrew Rabinovich, Alexander C Berg . arXiv:1506.04579, 2015

Hengshuang Zhao, Jianping Shi, Xiaojuan Qi, Xiaogang Wang, Jiaya Jia . In CVPR, 2017

Sergey Ioffe, Christian Szegedy . In ICML, 2015

Mark Sandler, Andrew Howard, Menglong Zhu, Andrey Zhmoginov, Liang-Chieh Chen . arXiv:1801.04381, 2018

François Chollet . In CVPR, 2017

Haozhi Qi, Zheng Zhang, Bin Xiao, Han Hu, Bowen Cheng, Yichen Wei, Jifeng Dai . ICCV COCO Challenge Workshop, 2017

Mark Everingham, S. M. Ali Eslami, Luc Van Gool, Christopher K. I. Williams, John Winn, and Andrew Zisserma. . IJCV, 2014

Cordts, Marius, Mohamed Omran, Sebastian Ramos, Timo Rehfeld, Markus Enzweiler, Rodrigo Benenson, Uwe Franke, Stefan Roth, Bernt Schiele. . In CVPR, 2016

Zhedong Zheng and Yi Yang, Unsupervised Scene Adaptation with Memory Regularization in vivo, IJCAI (2020)

Zhedong Zheng and Yi Yang, Rectifying Pseudo Label Learning via Uncertainty Estimation for Domain Adaptive Semantic Segmentation, IJCV (2020)

Alexander Kolesnikov, Christoph Lampert, Seed, Expand and Constrain: Three Principles for Weakly-Supervised Image Segmentation, ECCV, 2016

Guosheng Lin, Chunhua Shen, Ian Reid, Anton van dan Hengel, Efficient piecewise training of deep structured models for semantic segmentation, arXiv:1504.01013. (1st ranked in VOC2012)

Guosheng Lin, Chunhua Shen, Ian Reid, Anton van den Hengel, Deeply Learning the Messages in Message Passing Inference, arXiv:1508.02108. (4th ranked in VOC2012)

Ziwei Liu, Xiaoxiao Li, Ping Luo, Chen Change Loy, Xiaoou Tang, Semantic Image Segmentation via Deep Parsing Network, arXiv:1509.02634 / ICCV 2015 (2nd ranked in VOC 2012)

Hyeonwoo Noh, Seunghoon Hong, Bohyung Han, Learning Deconvolution Network for Semantic Segmentation, arXiv:1505.04366. (7th ranked in VOC2012)

Seunghoon Hong, Hyeonwoo Noh, Bohyung Han, Decoupled Deep Neural Network for Semi-supervised Semantic Segmentation, arXiv:1506.04924

Seunghoon Hong,Junhyuk Oh, Bohyung Han, and Honglak Lee, Learning Transferrable Knowledge for Semantic Segmentation with Deep Convolutional Neural Network, arXiv:1512.07928 [ ] [ ]

Liang-Chieh Chen, George Papandreou, Kevin Murphy, Alan L. Yuille, Weakly-and semi-supervised learning of a DCNN for semantic image segmentation, arXiv:1502.02734. (9th ranked in VOC2012)

Abhishek Sharma, Oncel Tuzel, David W. Jacobs, Deep Hierarchical Parsing for Semantic Segmentation, CVPR, 2015

Vijay Badrinarayanan, Alex Kendall and Roberto Cipolla "SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation." arXiv preprint arXiv:1511.00561, 2015

Fisher Yu, Vladlen Koltun, "Multi-Scale Context Aggregation by Dilated Convolutions", ICLR 2016, [ ]

Hamid Izadinia, Fereshteh Sadeghi, Santosh Kumar Divvala, Yejin Choi, Ali Farhadi, "Segment-Phrase Table for Semantic Segmentation, Visual Entailment and Paraphrasing", ICCV, 2015, [ ]

Iasonas Kokkinos, "Pusing the Boundaries of Boundary Detection Using deep Learning", ICLR 2016, [ ]

Niloufar Pourian, S. Karthikeyan, and B.S. Manjunath, "Weakly supervised graph based semantic segmentation by learning communities of image-parts", ICCV, 2015, [ ]

Unsupervised Monocular Depth Estimation with Left-Right Consistency [ ]

Deep Ordinal Regression Network for Monocular Depth Estimation [ ]

GeoNet: Unsupervised Learning of Dense Depth, Optical Flow and Camera Pose [ ]

Visual SLAM algorithms: a survey from 2010 to 2016[ ]

Visual map matching and localization using a global feature map [ ]

Map-based precision vehicle localization in urban environments [ ]

Simultaneous Localization And Mapping: A Survey of Current Trends in Autonomous Driving [ ]

Past, Present, and Future of Simultaneous Localization and Mapping: Toward the Robust-Perception Age [ ]

The GraphSLAM Algorithm with Applications to Large-Scale Mapping of Urban Structures [ ]

Large-scale mapping in complex field scenarios using an autonomous car [ ]

Road-SLAM : Road Marking based SLAM with Lane-level Accuracy [ ]

LIMO: Lidar-Monocular Visual Odometry [ ]

Visual-lidar Odometry and Mapping: Low-drift, Robust, and Fast [ ]

LOAM: Lidar Odometry and Mapping in Real-time [ ]

Efficient Surfel-Based SLAM using 3D Laser Range Data in Urban Environments [ ]

SOFT-SLAM: Computationally Efficient Stereo Visual SLAM for Autonomous UAVs [ ]

Visual SLAM for Automated Driving: Exploring the Applications of Deep Learning [ ]

SegMap: 3D Segment Mapping using Data-Driven Descriptors [ ]

SegMatch: Segment Based Place Recognition in 3D Point Clouds [ ]

Incremental Segment-Based Localization in 3D Point Clouds [ ]

Direct Visual SLAM using Sparse Depth for Camera-LiDAR System [ ]

Review of visual odometry:types, approaches, challenges, and applications [ ]

Semantic segmentation-aided visual odometry for urban autonomous driving [ ]

Vision-based ACC with a Single Camera : Bounds on Range and Range Rate Accuracy [ ]

Towards End-to-End Lane Detection: an Instance Segmentation Approach[ ]

Vision-Based Lane Analysis: Exploration of Issues and Approaches for Embedded Realization [ ]

Drive Analysis Using Vehicle Dynamicsand Vision-Based Lane Semantics [ ]

Planning and Decision-Making for Autonomous Vehicles[ ]

Perception, planning, control, and coordination for autonomous vehicles [ ]

A survey of motion planning and control techniques for self-driving urban vehicles [ ]

Real-time motion planning methods for autonomous on-road driving: State-of-the-art and future research directions [ ]

Behavior and path planning algorithm of autonomous vehicle A1 in structured environments [ ]

How Does Path Planning for Autonomous Vehicles Work [ ]

Towards full automated drive in urban environments: A demonstration in gomentum station, california [ ]

A behavioral planning framework for autonomous driving [ ]

Towards a functional system architecture for automated vehicles [ ]

Autonomous Driving: Planning, Control & Other Topics. (UNC presentation slides) [ ]

Udacity Self-Driving Car Nano Degree program description [ ]

Optimal trajectory generation for dynamic street scenarios in a frenet frame[ ]

Path planning for autonomous vehicles in unknown semi-structured environments[ ]

Local path planning for off-road autonomous driving with avoidance of static obstacles[ ]

Trajectory planning for Bertha—A local, continuous methods[ ]

Efficient sampling-based motion planning for on-road autonomous driving[ ]

Real-time motion planning methods for autonomous on-road driving: State-of-the-art and future research directions[ ]

A Review of Motion Planning Techniques for Automated Vehicles[ ]

A survey of motion planning and control techniques for self-driving urban vehicles[ ]

Real-time trajectory planning for autonomous urban driving: Framework, algorithms, and verifications[ ]

Dynamic path planning for autonomous driving on various roads with avoidance of static and moving obstacles[ ]

Hybrid Trajectory Planning for Autonomous Driving in Highly Constrained Environments[ ]

Vehicle path planning in various driving situations based on the elastic band theory for highway collision avoidance[ ]

Tasks of interest are: stereo, optical flow, visual odometry, 3D object detection and 3D tracking. For this purpose, we equipped a standard station wagon with two high-resolution color and grayscale video cameras. Accurate ground truth is provided by a Velodyne laser scanner and a GPS localization system

Large-scale dataset that contains a diverse set of stereo video sequences recorded in street scenes from 50 different cities, with high quality pixel-level annotations of 5 000 frames in addition to a larger set of 20 000 weakly annotated frames. Focused on developing Pixel Level Classification, Instance-wise Segmentation

A diverse street-level imagery dataset with pixel‑accurate and instance‑specific human annotations for understanding street scenes around the world. 25,000 high-resolution images,152 object categories,100 instance-specifically annotated categories,Global reach, covering 6 continents, Variety of weather, season, time of day, camera, and viewpoint

Scene Parsing ,Car Instance,Lane Segmentation,Self Localization,Trajectory

SYNTHIA, The SYNTHetic collection of Imagery and Annotations, is a dataset that has been generated with the purpose of aiding semantic segmentation and related scene understanding problems in the context of driving scenarios. SYNTHIA consists of a collection of photo-realistic frames rendered from a virtual city and comes with precise pixel-level semantic annotations

-The nuScenes dataset (pronounced /nuːsiːnz/) is a public large-scale dataset for autonomous driving provided by nuTonomy-Aptiv. By releasing a subset of our data to the public, we aim to support public research into computer vision and autonomous driving. For this purpose we collected 1000 driving scenes in Boston and Singapore, two cities that are known for their dense traffic and highly challenging driving situations. The scenes of 20 second length are manually selected to show a diverse and interesting set of driving maneuvers, traffic situations and unexpected behaviors. The rich complexity of nuScenes will encourage development of methods that enable safe driving in urban areas with dozens of objects per scene. Gathering data on different continents further allows us to study the generalization of computer vision algorithms across different locations, weather conditions, vehicle types, vegetation, road markings and left versus right hand traffic

The Daimler Urban Segmentation Dataset consists of video sequences recorded in urban traffic. The dataset consists of 5000 rectified stereo image pairs with a resolution of 1024x440. 500 frames (every 10th frame of the sequence) come with pixel-level semantic class annotations into 5 classes: ground, building, vehicle, pedestrian, sky. Dense disparity maps are provided as a reference, however these are not manually annotated but computed using semi-global matching (sgm)

Dataset for Traffic Sign Classification, Traffic Sign Detection

Traffic Lights Recognition (TLR) public benchmarks

The Caltech Pedestrian Dataset consists of approximately 10 hours of 640x480 30Hz video taken from a vehicle driving through regular traffic in an urban environment. About 250,000 frames (in 137 approximately minute long segments) with a total of 350,000 bounding boxes and 2300 unique pedestrians were annotated. The annotation includes temporal correspondence between bounding boxes and detailed occlusion labels

Caltech Lanes dataset includes four clips taken around streets in Pasadena, CA at different times of day. The archive below inlucdes 1225 individual frames as taken from a camera mounted on Alice in addition to the labeled lanes. The dataset is divided into four individual clips: cordova1 with 250 frames, cordova2 with 406 frames, washington1 with 337 frames, and washington2 with 232 frames

Udacity

ROSBAG training data. (~80 GB)

This data set provides Light Detection and Ranging (LiDAR) data with various position sensors targeting a highly complex urban environment. The presented data set captures features in urban environments (e.g. metropolis areas, complex buildings and residential areas). The data of 2D and 3D LiDAR are provided, which are typical types of LiDAR sensors. Raw sensor data for vehicle navigation is presented in a file format. For convenience, development tools are provided in the Robot Operating System (ROS) environment

The Oxford RobotCar Dataset contains over 100 repetitions of a consistent route through Oxford, UK, captured over a period of over a year. The dataset captures many different combinations of weather, traffic and pedestrians, along with longer term changes such as construction and roadworks

Here, you can find two challenging datasets recorded with the Velodyne HDL64E-S2 scanner in the city of Karlsruhe, Germany

long-term autonomy dataset for robotics research collected on the University of Michigan’s North Campus. The dataset consists of omnidirectional imagery, 3D lidar, planar lidar, GPS, and proprioceptive sensors for odometry collected using a Segway robot. The dataset was collected to facilitate research focusing on longterm autonomous operation in changing environments. The dataset is comprised of 27 sessions spaced approximately biweekly over the course of 15 months. The sessions repeatedly explore the campus, both indoors and outdoors, on varying trajectories, and at different times of the day across all four seasons. This allows the dataset to capture many challenging elements including: moving obstacles (e.g., pedestrians, bicyclists, and cars), changing lighting, varying viewpoint, seasonal and weather changes (e.g., falling leaves and snow), and long-term structural changes caused by construction projects

Dataset collected by an autonomous ground vehicle testbed, based upon a modified Ford F-250 pickup truck. The vehicle is outfitted with a professional (Applanix POS LV) and consumer (Xsens MTI-G) Inertial Measuring Unit (IMU), a Velodyne 3D-lidar scanner, two push-broom forward looking Riegl lidars, and a Point Grey Ladybug3 omnidirectional camera system. Here we present the time-registered data from these sensors mounted on the vehicle, collected while driving the vehicle around the Ford Research campus and downtown Dearborn, Michigan during November-December 2009. The vehicle path trajectory in these datasets contain several large and small-scale loop closures, which should be useful for testing various state of the art computer vision and SLAM (Simultaneous Localization and Mapping) algorithms. The size of the dataset is huge (~100 GB) so make sure that you have sufficient bandwidth before downloading the dataset

The dataset was recorded by placing a HD camera in a car driving around the Surrey countryside. The dataset contains about 30 minutes of driving. The video is 1920x1080 in colour, encoded using H.264 codec. Steering is estimated by tracking markers on the steering wheel. The car's speed is estimated from OCR the car's speedometer (but the accuracy of the method is not guaranteed)

7 and a quarter hours of largely highway driving. Consists of 10 videos clips of variable size recorded at 20 Hz with a camera mounted on the windshield of an Acura ILX 2016. In parallel to the videos, also recorded some measurements such as car's speed, acceleration, steering angle, GPS coordinates, gyroscope angles. These measurements are transformed into a uniform 100 Hz time base. color and grayscale stereo cameras, a Velodyne 3D laser scanner and a high-precision GPS/IMU inertial navigation system

1,000+ hours of multi-sensor driving datasets collected at AgeLab(Lex Fridman)

A large dataset with traffic sign annotations, thousands of physically distinct traffic signs

traffic sign, vehicles detection, traffic lights, trajectory patterns

Datasets drive vision progress and autonomous driving is a critical vision application, yet existing driving datasets are impoverished in terms of visual content. Driving imagery is becoming plentiful, but annotation is slow and expensive, as annotation tools have not kept pace with the flood of data. Our first contribution is the design and implementation of a scalable annotation system that can provide a comprehensive set of image labels for large-scale driving datasets. Our second contribution is a new driving dataset, facilitated by our tooling, which is an order of magnitude larger than previous efforts, and is comprised of over 100K videos with diverse kinds of annotations including image level tagging, object bounding boxes, drivable areas, lane markings, and full-frame instance segmentation. The dataset possesses geographic, environmental, and weather diversity, which is useful for training models so that they are less likely to be surprised by new conditions. The dataset can be requested at this http URL

Virtual KITTI is a photo-realistic synthetic video dataset designed to learn and evaluate computer vision models for several video understanding tasks: object detection and multi-object tracking, scene-level and instance-level semantic segmentation, optical flow, and depth estimation. Virtual KITTI contains 50 high-resolution monocular videos (21,260 frames) generated from five different virtual worlds in urban settings under different imaging and weather conditions. These worlds were created using the Unity game engine and a novel real-to-virtual cloning method. These photo-realistic synthetic videos are automatically, exactly, and fully annotated for 2D and 3D multi-object tracking and at the pixel level with category, instance, flow, and depth labels

Bosch Small Traffic Lights Dataset, an accurate dataset for vision-based traffic light detection. Vision-only based traffic light detection and tracking is a vital step on the way to fully automated driving in urban environments. We hope that this dataset allows for easy testing of objection detection approaches, especially for small objects in larger images

Dataset for Belgium Traffic Sign Classification, Detection

In contrast to Europe and the USA, most TLs for vehicles in South Korea at intersections have a horizontal layout and are installed as side-pillar horizontal types. A TL can have three or four signals and one signal consists of a 355 mm x 355 mm black box with colored bulbs. The diameter of the bulb is 300 mm. There are two types of bulbs: a circle and an arrow. The circle bulb indicates green, red, and yellow, whereas the arrow bulb represents a left turn. There are two combinations for the three bulb TL, and there is one type for the four bulb TL. The TL status can be green, yellow, red, green + left turn, and red + left turn

Provides Millimetre-Wave radar data, dual velodyne lidars, and optimised ground truth odometry for 280 km of driving around Oxford, UK (in addition to all sensors in the original )

[Stanford]

[CUHK]

[Oxford]

[NYU]

ROS:he Robot Operating System (ROS) is a set of software libraries and tools that help you build robot applications. From drivers to state-of-the-art algorithms, and with powerful developer tools, ROS has what you need for your next robotics project. And it's all open source. [ ]

Tensorflow: An open source software library for numerical computation using data flow graph by Google [ ]

PyTorch: Deep learning library in Python, used by Facebook [ ]

Torch7: Deep learning library in Lua, used by Facebook and Google Deepmind [ ]

torchnet

Torch-based deep learning libraries: [ ],

Caffe: Deep learning framework by the BVLC [ ]

Caffe2: Deep learning framework by the Facebook [ ]

MXNet: A flexible and efficient deep learning library for heterogeneous distributed systems with multi-language support [ ]

Keras: The Python Deep Learning library [ ]

CNTK : The Microsoft Cognitive Toolkit [ ]

Chainer : Python-based deep learning framework for neural networks that is designed by the run strategy [ ]

Theano: Mathematical library in Python, maintained by LISA lab [ ]

Theano-based deep learning libraries: [ ], [ ], [ ], [ ]

MatConvNet: CNNs for MATLAB [ ]

[CVPR 2018 Main Conferece]

[CVPR 2018 Tutorial]

[CVPR 2018 Workshop]

[ICML IJCAI 2018]