Skip to main content

CLI Downloader

info

As a pre-requisite, you must:

  1. Make sure you have a License Agreement. If you don't, please refer to the Getting Started page. The License Agreement process takes 2 days: so do it early!
  2. Have setup or installed the Ego4D codebase, please refer to the Ego4D Github README.
danger

If you have downloaded the data prior to December 15th (Friday), 2023: 5PM (PST) and/or have files downloaded in <download_dir>/takes/takes or <download_dir/captures/captures this is due to a bug. These files should live in <download_dir>/takes/ or <download_dir>/captures/.

Here is a script to correct this, please also re-download with -d. Apologies for the inconvenience.

The dataset is large, due to this it is partitioned into separate "parts" (subsets). You may choose to download one or more parts of the dataset with optional filters in order to reduce how much you need to download.

The different parts of the dataset include:

  • metadata
  • annotations
  • takes
  • captures
  • take_trajectory
  • take_eye_gaze
  • take_vrs_noimagestream
  • take_point_cloud
  • take_vrs
  • capture_trajectory
  • capture_eye_gaze
  • capture_point_cloud
  • downscaled_takes/448
  • features/omnivore_video

The parts in bold refer to the recommended set. You can additionally filter based on:

  • Data relevant to a specific benchmarks: --benchmarks <benchmark_name>
    • e.g. --benchmarks keystep relations
  • Dataset splits (train/val/test): --splits <s1> <s2> ..., e.g.
    • --splits train for data that only lies in the training set
  • Egocentric or exocentric related data (videos, trajectory)
    • Use --views exo for only exocentric, --views ego for only egocentric, or --views ego exo for both. Defaults to both.
  • Specific takes or captures to download: --uids
  • Universities (use --universities or -u)
    • e.g. -u cmu unc sfu

Please refer to the Download CLI README or use the flag --help for the latest parts of the dataset and additional usage of the dataset.

Download The Data

Below are some examples on how to run the downloader CLI with common options:

As noted above, you can use --help to view all the up-to-date options
egoexo --help

Download suggested dataset parts for all the takes present.

This will download ~14TiB of data
egoexo -o <out-dir>

Other Examples

Download all annotations & metadata:

egoexo -o <out-dir> --parts annotations metadata

Download all point clouds, eye gaze and trajectories for each take:

egoexo -o <out-dir> --parts take_point_cloud take_eye_gaze trajectory

Download all point clouds, eye gaze and trajectories:

egoexo -o <out-dir> --parts take_point_cloud eye_gaze take_trajectory