Event-Based Visual Place Recognition With Ensembles of Temporal Windows

This dataset accompanies the following publication, please cite this publication if you use this dataset:

Fischer, T. and Milford, M., 2020. Event-Based Visual Place Recognition With Ensembles of Temporal Windows. IEEE Robotics and Automation Letters, 5(4), pp.6924-6931.

@article{fischer2020event,
  title={Event-Based Visual Place Recognition With Ensembles of Temporal Windows},
  author={Fischer, Tobias and Milford, Michael},
  journal={IEEE Robotics and Automation Letters},
  volume={5},
  number={4},
  pages={6924--6931},
  year={2020}
}

The dataset contains six sequences of recordings. For each recording, four files are made available:

1. A rosbag (*.bag) file with the following contents:

   • /dvs/events (type: dvs_msgs/EventArray) with the event stream, see https://github.com/uzh-rpg/rpg_dvs_ros)

   • /dvs/camera_info (type: sensor_msgs/CameraInfo) with the camera info of the DAVIS frame camera

   • /dvs/image_raw (type: sensor_msgs/Image) with the DAVIS frame camera images

   • /dvs/imu (sensor_msgs/Imu) with the IMU data of the event camera

2. An associated *_hot_pixels.txt file that contains the hot pixels for that recording (detected with https://github.com/cedric-scheerlinck/dvs_tools/tree/master/dvs_hot_pixel_filter)

3. The recordings using a frame-based consumer camera (*.mp4 files)

4. Associated GPS information (*.nmea) files recorded using the consumer camera - synchronized with the mp4 files.

Note that the timestamps of the event files and frame-based camera files do not match as they were not synchronized during recording time. Please see the associated code repository (https://github.com/Tobias-Fischer/ensemble-event-vpr) for the correct mappings and time offsets between the event recordings and frame-based recordings.

The code repository also contains code to match GPS data between different traverses.