Geppetto tracks 1035 robot training datasets including Open X-Embodiment, DROID, and AgiBot World, each with structured metadata on trajectory count, hours of demonstration, robot platforms, license, commercial use, access model, and file format. Every dataset connects to the robots it trains and the professions it affects.

DatasetYearCategoryTrajectoriesHoursPlatformsLicenseCommercialAccessFormat
Boreas2022research-lab——car with Velodyne Alpha Prime, Navtech radar, FLIR cameranot statedNoopen—
CODA2022research-lab——images from KITTI, nuScenes, ONCECC BY-NC-SA 4.0Noopen—
DAIR-V2X2022research-lab——vehicle and roadside cameras and lidarnot statedNogated—
DualCam2022research-lab——dual camera traffic lightsnot statedNoopen—
IDD-3D2022research-lab——car with cameras and lidarnot statedNogated—
Ithaca3652022research-lab——car with cameras and lidarCC BY-NC-SA 4.0Noopen—
K-Lane2022research-lab——lidar and cameraApache-2.0Yesopen—
K-Radar2022research-lab——car with 4D radar, lidar, stereo camerasCC BY-NC-NDNoopen—
ONCE-3DLanes2022research-lab——ONCE vehiclenot statedNogated—
OpenLane2022research-lab——Waymo Open images with 3D lanesnot statedNoopen—
PSI2022research-lab——dashcam videonot statedNoopen—
SceNDD2022research-lab——naturalistic drivingnot statedNoopen—
TJ4DRadSet2022research-lab——car with 4D radar, lidar, cameranot statedNoopen—
View-of-Delft2022research-lab——car with 64-layer lidar, stereo camera, 3+1D radarnot statedNoopen—
WildDash 22022research-lab——dashcam images worldwideregistration requiredNogated—
ActionSense2022human-video——eye tracker with first-person camera, EMG, 17-IMU suit, tactile glovesnot statedNoopen—
ARCTIC2022human-video——mocap studio with static cameras and an egocentric cameranot statedNoopen—
AssistQ2022human-video——newly filmed egocentric instructional videosnot statedNoopen—
EDINA2022human-video——Egocentric RGB with depth and surface normalsnot statedNoopen—
EgoHOS2022human-video——frames from several egocentric setsMIT (per atlas)Yesopen—
EgoTaskQA2022human-video——LEMMA egocentric videosnot statedNoopen—
FT-HID2022human-video——RGB-D first- and third-person capturenot statedNoopen—
GIMO2022human-video——head-mounted device with gaze, inertial motion capturenot statedNoopen—
HOI4D2022human-video——head-mounted RGB-DCC BY-NC 4.0 (per atlas)Noopen—
SceneEgo (EgoGTA, EgoPW-Scene)2022human-video——head-mounted fisheye camera, synthetic and realnot statedNoopen—
UmeTrack2022human-video——VR headset with 4 fisheye sensorsCC BY-NC 4.0Noopen—
VOST2022human-video——egocentric videos from existing setsnot statedNoopen—
OakInk2022human-video——multi-camera third-person rig (not egocentric)CC-BY-NC-SA-3.0 (Hub oakink/OakInk-v1)Noopen—
UnrealEgo / UnrealEgo2 / UnrealEgo-RW2022human-video——virtual eyeglasses with two fisheye cameras; RW adds real devicenot statedNoopen—
DA22022research-lab——dual-arm parallel-jaw grasp pairsnot statedNoopen—
DexGraspNet2022research-lab——ShadowHand, validated in Isaac GymCC BY-NC 4.0Noopen—
MultiDex (GenDexGrasp)2022research-lab——multiple robotic handsnot statedNoopen—
DREDS and STD2022research-lab——simulated and real RGB-D of specular, transparent, diffuse objectsCC BY-NC 4.0Noopen—
TransCG2022research-lab——RGB-D of transparent objects, robot-collectedCC BY-NC-SA 4.0Noopen—
RT-1 Robot Action Dataset2022human-video—350.6—CC-BY-4.0Yesopen—
CALVIN2022synthetic—24franka_pandaMITYesopen—
Language-Table2022real-robot—3,865xarmApache-2.0Yesopen—
Ego4D2022human-video—3,670humanresearch-onlyNoopen—
EPIC-KITCHENS-1002022human-video—100humanCC-BY-NC-4.0Noopen—
TartanDrive2022human-video—5offroad_vehicleCC-BY-4.0Yesopen—
ALTO2022research-lab——helicopter with downward camera, GPS-INS, laser altimeternot statedNoopen—
DenseUAV2022research-lab——low-altitude UAV plus satellitenot statedNoopen—
HIT-UAV2022research-lab——UAV infrared thermalnot statedNoopen—
INSANE2022research-lab——multi-sensor MAVnot statedNoopen—
NTU VIRAL2022research-lab——UAV with two 3D lidars, two cameras, IMUs, UWBnot statedNoopen—
Rope3D2022research-lab——roadside cameras with lidar labelsregistration required (per ad-datasets)Nogated—
MetaGraspNet / MetaGraspNetV22022research-lab——synthetic and real RGBD bin scenes, parallel-jaw and vacuum labelssee repo License.txtNoopen—
PoseIt2022research-lab——robot arm grasping with GelSight and force sensing, RGB-Dnot statedNoopen—
BEHAVE2022research-lab——human body plus object, multi-view RGBD, SMPL fitsnot statedNoopen—
COUCH2022research-lab——human body with chairsnot statedNoopen—
FLAG3D2022research-lab——human fitness activity, mocap plus rendered and real videonot statedNoopen—
HODome (NeuralDome)2022research-lab——human plus object in dense multi-view domenot statedNoopen—
HuMMan2022research-lab——human body, RGB-D, point clouds, SMPL, textured meshesnot statedNoopen—
InterCap2022research-lab——human plus object, 6 RGB-D camerasnot statedNoopen—
RICH2022research-lab——human bodies with scene scans, multi-view 4K videonot statedNoopen—
ZeroEGGS2022research-lab——human full body gesture with fingers and speechnot statedNoopen—
ASE sword and shield motions2022research-lab——simulated humanoid characternot statedNoopen—
CHICO2022research-lab——human and cobot in industrial tasksnot statedNoopen—
SQA3D2022research-lab——650 ScanNet scenesCC-BY-4.0 (data), Apache (code)Noopen—
DialFRED2022research-lab——AI2-THORnot statedNoopen—
Housekeep2022research-lab——Habitat householdsnot statedNoopen—
IVLN (Iterative VLN)2022research-lab——Matterport3D and Habitatnot statedNoopen—
DRAMA2022research-lab——vehicle camera, Tokyoresearch onlyNoopen—
AMP for hardware A1 mocap motions2022research-lab——Unitree A1not statedNoopen—
Cerberus visual-inertial-leg odometry data2022research-lab——Unitree quadrupedsnot statedNoopen—
UTIL2022research-lab——Decawave DWM1000 UWB modulesnot statedNoopen—
FusionPortable2022research-lab——portable multi-sensor suite on diverse platformsnot statedNoopen—
OdomBeyondVision2022research-lab——UAV, UGV and handheld with radar, thermal and other sensorsnot statedNoopen—
EVIMO22022research-lab——event camerasnot statedNoopen—
GSV-Cities2022research-lab——Google Street View place imagesnot statedNoopen—
Hilti-Oxford Dataset (2022)2022research-lab——handheld LiDAR, camera, IMU rignot statedNoopen—
Multi-Modal Lidar Dataset (general-purpose localization and mapping)2022research-lab——multiple lidar typesnot statedNoopen—
T2FPV2022research-lab——first-person pedestrian trajectoriesnot statedNoopen—
VECtor2022research-lab——event-centric multi-sensor rignot statedNoopen—
ViViD++2022research-lab——thermal, event, RGB-D, lidar sensorsnot statedNoopen—
ALITA2022research-lab——LiDAR and omnidirectional cameranot statedNoopen—
S3E2022research-lab——fleet of unmanned ground vehiclesnot statedNoopen—
TGRMPT2022research-lab——tour-guide robotnot statedNoopen—
TorWIC (Mapping and SLAM)2022research-lab——mobile robot in Clearpath Robotics warehousenot statedNoopen—
RoboMimic — A Study of Imitation Learning from Observations2021real-robot——Franka Panda, SawyerMITYesopenhdf5
SeaDronesSee2021research-lab——UAV over open waternot statedNoopen—
VID (Visual-Inertial-Dynamical) Dataset2021research-lab——multirotor with rotor speed and force sensingnot statedNoopen—
Automatum Open Highway2021research-lab——drone over German highwaysCC BY-NDNoopen—
Topo-boundary2021research-lab——aerial images of road boundariesnot statedNoopen—
ACDC2021research-lab——car camera, Switzerlandnot statedNogated—
DSEC2021research-lab——stereo event cameras, RGB, lidar, GPSnot statedNoopen—
DurLAR2021research-lab——car with 128-channel lidar, stereo camera, lux meter, GNSS/INSnot statedNoopen—
Honda LOKI2021research-lab——vehicle camera and lidarresearch onlyNoopen—
KITTI-3602021research-lab——car with fisheye and perspective cameras, lidarCC BY-NC-SA 3.0Noopen—
Oxford Road Boundaries2021research-lab——Oxford RobotCarnot statedNoopen—
PixSet2021research-lab——car with full-waveform flash lidar, cameras, radarnot statedNoopen—
PVDN (Provident Vehicle Detection at Night)2021research-lab——grayscale camera at nightnot statedNoopen—
RADIal2021research-lab——car with HD radar, camera, laser scannernot statedNoopen—
ROAD2021research-lab——Oxford RobotCar videosCC BY-SA 4.0Yesopen—
SegmentMeIfYouCan (RoadAnomaly21, RoadObstacle21)2021research-lab——road images with anomaliesvariousNoopen—
Shifts (vehicle motion prediction)2021research-lab——Yandex self-driving fleet tracksnot statedNoopen—
SODA10M2021research-lab——dashcam imagesCC BY-NC-SA 4.0Nogated—
H2O2021human-video——synchronised multi-view RGB-D including egocentricnot statedNoopen—
HPS (Human POSEitioning System)2021research-lab——human body, body-worn IMUs plus head camera in scanned scenesnot statedNoopen—
N-EPIC-Kitchens2021human-video——EPIC-KITCHENS videos converted to event streamsnot statedNoopen—