# Data Format of Trajectory in Future Localization Dataset

**URL:** <https://discuss.ego4d-data.org/t/data-format-of-trajectory-in-future-localization-dataset/693>\
**Category:** Q&A\
**Tags:** forecasting\
**Created:** [September 5, 2024, 6:03pm UTC](https://discuss.ego4d-data.org/t/data-format-of-trajectory-in-future-localization-dataset/693 "2024-09-05T18:03:02Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![george](https://avatars.discourse-cdn.com/v4/letter/g/71e660/32.png) [@george](https://discuss.ego4d-data.org/u/george)\
**Post date:** [September 5, 2024, 6:03pm UTC](https://discuss.ego4d-data.org/t/data-format-of-trajectory-in-future-localization-dataset/693/1 "2024-09-05T18:03:02Z")

</div>

Hello,

Can someone please explain the data format of the trajectory in the Future Localization dataset? There is no information on what t\_n or b is in the example provided below:

[up.x up.y up.z] TrajLength(n) [t\_0 C\_0.x C\_0.y C\_0.z b b] [t\_1 C\_1.x C\_1.y C\_1.z b b] … [t\_n …]

Additionally, could someone explain how the trajectory relates to the camera frame?

Thanks in advance!