Home

Awesome

Awesome SLAM Datasets Awesome

image

Thumbnail Figures from Complex Urban, NCLT, Oxford robotcar, KiTTi, Cityscapes datasets.

This repository is the collection of SLAM-related datasets. Among various SLAM datasets, we've selected the datasets provide pose and map information. This repository is linked to the google site. In this repository, the overall dataset chart is represented as simplified version. You can use full version of the chart (made by google spreadsheet) in the project page.

We provide several category for each access of the data.

News !!

Update: 2024-07-31

Update: 2024-4-18

Update: 2023-06-20

Update: 2022-8-15

Update: 2022-04-07

Update: 2021-03-24

Update: 2021-02-26

Update: 2020-02-29

Update: 2019-09-24

TODO

Category

Overall datasets chart (Simplified Version)

Link to Full version

ShortnameAffiliationYearPlatformPublicationEnvironmentGT-PoseGT-MapIMUGPSLabelsLidarCamerasRGBDEventRadarSonarDVLOther
VBR SLAM DatasetSapienza University of Rome2024Hand, VehICRAIndoor + OutdoorOOOOStereo Cams; Multiple configuration; SLAM Benchmark
4Seasons DatasetTechnical University of Munich2020VehGCPROutdoorOOOOOOOO
M2DGRShanghai Jiaotong University2021MobRA-LIndoor + OutdoorOOOOOOO
TartanAirCMU2020UAVIROSSimulationOOOOO
VECtor DatasetShanghaiTech University2022HandRA-LIndoor + OutdoorOOOOOO
Hilti SLAM DatasetHilti, Oxford, UZH2022HandIndoor + OutdoorOOOOO
MADMAX Mars DatasetGerman Aerospace Center (DLR)2021Handheld Rover-MockupJFRRough Terrain, Planetary analogOOOOOStereo Cams + Stereo Omnidirectional Cameras; Navigation Benchmark
FinnForest DatasetTampere University2020VehRAS journalForest, TerrainOOMultiple Stereo Pairs ; Sampling Rate(40, 13, and 8Hz)
PennCOSYVIO DatasetUniversity of Pennsylvania2017HandICRAIndoor + OutdoorOOO
ICL DatasetImperial College2019Hand, MAVICRAIndoorOOO
UZH-FPV Drone RacingUZH, ETH2019UAVICRAIndoor, UrbanOOOO
FMDatasetHangzhou Dianzi / Tsinghua2019HandICMEIndoorOO
Rosario DatasetCONICET-UNR2019MobIJRRTerrainOOOEncoder
Collaborative SLAM Dataset (CSD)Oxford2018HandTVCG/ISMARIndoorOOOOOTango (Asus ZenFone AR)
ADVIO DatasetAalto U2018HandECCVUrbanOOOOiPhone, Tango, Pixel
DeepIO DatasetOxford2018HandArxivIndoorOO
Aqualoc DatasetONERA-DTIS2018ROVIROS WSUnderwaterOOOPressure Sensor
InteriorNetImperial College2018HandBMVCIndoorOOOOOOOTexture, Lighting, Context, Optical Flow
SPO DatasetTUM, Karlsruhe2018HandArxivUrbanOOPlenoptic Camera
Complex UrbanKAIST-IRAP2018VehICRAUrbanOOOOOEncoder
KAIST Day/NightKAIST-RCV2018VehT-ITSUrbanOOOOOOThermal Camera
TUM-Visual-InertialTUM2018HandArxivIndoor, UrbanOOO
Multi Vech EventUpenn2018VehRA-LUrbanOOOOOO
VI CanoeUIUC2018USVIJRRTerrainOOOO
MPO-JapanETH-RPG2017UAV / HandIJRRIndoorOOOO
Underwater CaveUDG2017AUVIJRRUnderwaterOOOOOProfiling Sonar
Robot @ HomeMRPT2017MobIJRRIndoorOOOOOSemantic Labels
Zurich Urban MAVETH-RPG2017UAVIJRRUrbanOOOOStreetview images
Chilean UndergroundTrimble2017MobIJRRTerrain (Underground)OOOOEncoder
SceneNet RGB-DImperial2017HandICCVIndoorOOO
Symphony LakeGeorgia Tech2017USVIJRRTerrain (Lake)OOOOPTZ camera, Longterm
Agricultural robotBonn2017MobIJRRTerrainOOOOOOMultispectral camera
Beach RoverTEC-MMA2017MobTerrainOOOOOOEncoder
EuRoCETH-ASL2016UAVIJRRIndoorOOOO
CartographerGoogle2016HandICRAIndoorOOO
TUM-MonoTUM2016HandArxivIndoor, UrbanOPhotometric Calibration
CityscapeDaimler AG2016VehCVPRUrbanOOOO
Solar-UAVETHZ2016UAVCVPRTerrainOOOOO
CoRBSDFKI2016HandWACVIndoorOOO
Oxford-robotcarOxford2016VehIJRRUrbanOOOOO
NCLTUMich2016MobIJRRUrbanOOOOFOG
RPG-eventKyushu U2016VehIROSUrban, TerrainOOOOFARO 3D
CCSADCIMAT2015VehCAIPUrbanOOO
TUM-OmniTUM2015HandIROSIndoor, UrbanO
Augmented ICL-NUIMRedwood2015HandCVPRIndoorOOO
Cambridge LandmarkCambridge2015HandICCVUrbanOOO
ICL-NUIMImperial2014HandICRAIndoorOOO
MRPT-MalagaMRPT2014VehARUrbanOOOO
KITTIKIT2013VehIJRRUrbanOOOOOO
Canadian PlanetaryUToronto2013MobIJRRTerrainOOOO (sensor)O
Microsoft 7 scenesMicrosoft2013HandCVPRIndoorOOO
SeqSLAMQUT2012VehICRAUrbanOO
ETH-challengingETH-ASL2012HandIJRRUrban, TerrainOOOOO
TUM-RGBDTUM2012Hand / MobIROSIndoorOOO
ASRL-Kagara-airborneUToronto2012UAVFSRTerrainOOO
Devon Island RoverUToronto2012MobIJRRTerrainOOOSunsensor, Inclinometer
ACFR MarineACFR2012AUVUnderwaterOOOOO
UTIAS Multi-RobotUT-IAS2011MobIJRRUrbanOO
Ford CampusUMich2011VehIJRRUrbanOOOOO
San franciscoStanford2011VehCVPRUrbanOOOOODMI
Annotated-laserNTU2011VehIJRRUrbanOOOO
MIT-DARPAMIT2010VehIJRRUrbanOOOOOO
St Lucia StereoUToronto2010VehACRAUrbanOOO
St Lucia Multiple TimesQUT2010VehICRAUrbanOO
MarulanACFR2010MobIJRRTerrainOOOOOOIR
COLDKTH2009HandIJRRIndoorOOOO
NewCollegeOxford-Robot2009MobIJRRUrbanOOOOO
Rawseeds-indoorMilano2009MobIROSWIndoorOOOOOO
Rawseeds-outdoorMilano2009MobIROSWUrbanOOOOOOO
FABMAPOxford-Robot2008VehIJRRUrbanOO

Evaluation

Evaluation methods for SLAM benchmarks

Categorized By Topic

Odometry

Dataset for odometry Benchmark

Mapping

Dataset for mapping task

Place Recognition

Dataset gives correspondences of places (images)

Localization

Dataset for metric-level localization

Perception

Dataset with semantic labels / correspondences

Categorized By Characteristics

Large-scale

City-scale map, kilometer level Map

Long-term

Multi-session, long-term data collection

Map Complexity

Variation of mapping structures

Extreme Condition

Extreme environment, motions

Categorized by Platform

Vehicle

Commercial Vehicle (Four-wheeled on the road)

Mobile Robot

Mobile Robots (Ex. Husky, Rover.. )

Unmanned Aerial Vehicle

Unmanned aerial robots include drone

Autonomous Underwater Vehicle

Underwater robots include ROV for simplicity

Unmanned Surface Vehicle

Water surface vehicle such as canoe and boat

Hand-held Device

Hand-held platform by human

Categorized by Environment

Urban

_City, campus, town, and infrastructures

Indoor

Indoor environment

Terrain

Rough terrain, underground, lake and farm

Underwater

Underwater floor, cave

Simulation

Simulation Scene

Contributing

Please Feel free to send a pull request to modify the list or add datasets.

License

CC0

To the extent possible under law, Younggun Cho has waived all copyright and related or neighboring rights to this work.