- Event-based Vision: A Survey, IEEE Trans. Pattern Anal. Machine Intell. (TPAMI), 2020.
-
From Real Scenes
-
DAVIS Data Dataset size:~10GB
paper name: The event-camera dataset and simulator: Event-based data for pose estimation, visual odometry, and SLAM, IJRR, 2017.
features : to motivate research on new algorithms for high-speed and high-dynamic-range, ground-truth camera poses from a motion-capture, a simulator that released open-source to create synthetic event-camera data.
task: SLAM
devices：DAVIS240C, motion-capture system -
EED Dataset size:59MB
paper name: Event-based Moving Object Detection and Tracking, IROS, 2018.
features : objects of multiple sizes moving at different speeds in a variety of lighting conditions indoor
task: moving object detection, moving object tracking
devices：DAVIS240B -
MVSEC Dataset size:~30G
paper name: The Multi Vehicle Stereo Event Camera Dataset: An Event Camera Dataset for 3D Perception, IEEE Robot. Autom. Lett.(RA-L), 2018.
features : a synchronized stereo pair event based camera system, carried on a handheld rig, flown by a hexacopter, driven on top of a car, and mounted on a motorcycle,in a variety of different illumination levels and environments
task: SLAM
devices：DAVIS m346B(stereo), VI-Sensor(stereo), Velodyne Puck LITE,GPS -
EV-IMO Dataset size:~30G
paper name: EV-IMO: Motion Segmentation Dataset and Learning Pipeline for Event Cameras, IROS, 2019.
features : 32 minutes of indoor recording with up to 3 fast moving objects in the camera field of view, including accurate pixel-wise motion masks, egomotion and ground truth depth. suitable for learning approach for motion segmentation in indoor scenes
task: motion segmentation
devices：DAVIS 346C -
DSEC Dataset size:~150G
paper name: DSEC: A Stereo Event Camera Dataset for Driving Scenarios, IEEE Robot. Autom. Lett.(RA-L),2021.
features :driving under challenging illumination conditions such as night, sunrise, and sunset
task: SLAM
devices: stereo event camera(2x Prophesee Gen3.1), GNSS, LiDAR, stereo RGB camera -
TUM-VIE Dataset size:~300G
paper name: TUM-VIE: The TUM Stereo Visual-Inertial Event Dataset, IROS,2021.
features : handheld and head-mounted sequences in indoor and outdoor environments, including rapid motion during sports and high dynamic range scenarios
task: SLAM
devices: stereo event camera(2x Prophesee Gen4-CD), motion capture, IMU, stereo RGB camera -
EventScape Dataset size:~70G
paper name: Combining Events and Frames Using Recurrent Asynchronous Multimodal Networks for Monocular Depth Prediction, RA-L,2021.
features : events, intensity frames, semantic labels, and depth maps recorded in the CARLA simulator
task: SLAM， semantic segmentation， depth estimation
devices: CARLA simulator -
ViViD++ Dataset size:more than 250min
paper name: ViViD++ : Vision for Visibility Dataset, RA-L,2022.
features : developing robust visual SLAM under poor illumination
devices1: mono event camera(Handheld DAVIS240C),Thermal,IMU,RGB-D
devices2: mono event camera(Driving DVXplorer),Thermal,RGB,RTK-GPS,Ouster OS1 LiDAR -
EV-IMO Dataset size:~40G
paper name: EVIMO2: An Event Camera Dataset for Motion Segmentation, Optical Flow, Structure from Motion, and Visual Inertial Odometry in Indoor Scenes with Monocular or Stereo Algorithms, arxiv, 2022.
features : temporal synchronization between sensors, less jitter, and a more efficient npz format
task: motion segmentation, SfM, Object Recognition
devices：2x Prophesee Gen3， Samsung Gen3, Prophesee ATIS Gen3
-
-
From Static Pictures
- N-Caltech101
- N-Cars
EED 202
DSEC 76
TUM-VIE 19
ViViD++ 12
Teams and researchers in China
Event-based Resources
Event-SLAM Hong Kong University