uFRED-NoSplit / README.md
JK-Ying-Long's picture
Update README.md
623eb45 verified
metadata
license: apache-2.0
dataset_info:
  features:
    - name: sequence_id
      dtype: int32
    - name: frame_id
      dtype: int32
    - name: timestamp
      dtype: string
    - name: rgb_image
      dtype: image
    - name: event_image
      dtype: image
    - name: bounding_box
      list: float32
    - name: track_id
      dtype: int32
    - name: class
      dtype: string
  splits:
    - name: train
      num_bytes: 230357706047
      num_examples: 1126513
  download_size: 148565226561
  dataset_size: 230357706047
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
pretty_name: Unzipped FRED No Splits

uFRED: Unzipped Florence RGB-Event Drone Dataset

This is a processed version of the FRED (Florence RGB-Event Drone Dataset) formatted for ease of use, with no data splits.

License

This dataset is licensed under Apache 2.0, matching the original FRED dataset license. This is a processed/reformatted version of the original FRED dataset. All credit for data collection goes to the original FRED authors. See citation below.

If you use this dataset, please cite the original FRED dataset:

@inproceedings{magrini2025fred,
title={FRED: The Florence RGB-Event Drone Dataset},
author={Magrini, Gabriele and Marini, Niccolò and Becattini, Federico and Berlincioni, Lorenzo and Biondi, Niccolò and Pala, Pietro and Del Bimbo,
Alberto},
booktitle={Proceedings of the 33rd ACM International conference on multimedia},
year={2025}
}