Skip to content
Multi-class Wild-life animal detection dataset
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Type Name Latest commit message Commit time
Failed to load latest commit information.

Multi-class Wild-life animal detection dataset

Multi-class Wild-life animal detection dataset

Evaluation Protocol:

Test: 40% Train: 20% Unlabelled: 40%

Evaluate maP with IOU 0.5


The Multiclass wildlife dataset (‘Licensed Material’) are made available to the non-commerical and scientific communities for academic, teaching and scientific purposes. Permission is granted to you (‘the Licensee’) to use the dataset with these conditions. Please include a reference to the following published work:

Teng Zhang, Liangchen Liu, Kun Zhao, Arnold Wiliem, Graham Hemson, Brian Lovell “Omni-supervised joint detection and pose estimation for wild animals” Pattern Recognition Letters, 2018


The current version of the Multiclass wildlife dataset can be downloaded from download

You can’t perform that action at this time.