KABR Raw Drone Videos of Giraffes and Zebras, Mpala Research Centre, Kenya

Provider
Imageomics Institute (Kline J. et al.)
License
CC0-1.0 (source)
Size
436 files
Format
MP4 (in sampled files)
Stored at
Hugging Face (https://huggingface.co/datasets/imageomics/KABR-raw-videos)
Access
Free and public. No account with the data provider is needed.
Category
Life Sciences
Source
https://arxiv.org/abs/2510.02030

About KABR Raw Drone Videos of Giraffes and Zebras, Mpala Research Centre, Kenya

The unprocessed drone footage behind KABR (Kenyan Animal Behavior Recognition), filmed at the Mpala Research Centre in Laikipia, Kenya, from 10 to 21 January 2023 with a DJI Air 2S at 4K to 5.4K and 29.97 fps. Videos follow reticulated giraffes, plains zebras and endangered Grevy's zebras in their habitat; some files were trimmed to remove take-off, landing, people or vehicles and are kept as _trimmed.mp4 beside the originals. Each video comes with its SRT flight log (GPS position, altitude, camera settings per frame), so footage can be tied to drone telemetry. Folders are organised by date and session (for example 18_01_2023_session_1 to 8), each with a readme.txt recording trimming. Measured: 436 files, 201 MP4 and 197 SRT, about 670 GB. Intended for animal detection, multi-object tracking, behaviour recognition pipelines and preprocessing tools such as kabr-tools, which extracts the behaviour mini-scenes released separately as imageomics/KABR. Dedicated to the public domain by the curators, who ask that you cite the KABR dataset and the kabr-tools paper (arXiv 2510.02030).

How to load this dataset in Python

from huggingface_hub import hf_hub_download
path = hf_hub_download(repo_id="imageomics/KABR-raw-videos", filename="11_01_23_session_1/DJI_0488.MP4", repo_type="dataset")
print(path)

Data structure

File types in sampled files: MP4

Files

Example file: 11_01_23_session_1/DJI_0488.MP4

Analyze this dataset without downloading it

On Scigantic, KABR Raw Drone Videos of Giraffes and Zebras, Mpala Research Centre, Kenya opens in a hosted JupyterLab notebook with the data already mounted, so you can read files where they are stored instead of copying them first. Each notebook includes an AI assistant that can see the mounted files and write and run analysis code. Open the dataset page to launch a notebook, or browse the full catalog.

This page is also available as Markdown and as JSON.

Related Life Sciences datasets