UAV Wildlife Detection Datasets: Kangaroo, Koala, and Waterbird

<b>1. Included Datasets</b>The <b>Kangaroo dataset</b> consists of thermal images captured at a resolution of 640×512 using drone-mounted sensors. It includes 1,149 training images and 289 validation images. The annotations cover three classes: <i>Kangaroo</i>, <i>Non-Arboreal</i> (representing canopy-level heat sources), and <i>Non-Kangaroo</i> (unidentified ground-level objects). Annotations are provided in both YOLO <code>.txt</code> format and COCO-style <code>.json</code> files.The <b>Koala dataset</b> also uses thermal imagery with a resolution of 640×512 and was collected in dense eucalyptus forest environments. The dataset comprises 4,277 training images and 1,252 validation images, annotated with two classes: <i>Koala</i> and <i>Non-Koala</i>. Similar to the Kangaroo dataset, annotations are available in both YOLO and COCO formats, enabling compatibility with a wide range of object detection frameworks.The <b>Waterbird dataset</b> contains high-resolution RGB imagery (5472×3648 and 8192×5460) collected in wetland environments. It includes 390 training images and 128 validation images. This dataset is annotated with ten fine-grained bird categories, including <i>Eurasian Coot</i>, <i>Chestnut Teal</i>, <i>Grey Teal</i>, and others. Annotations are provided in both YOLO and COCO formats, supporting multi-class and small object detection in complex natural scenes.<b>2. Folder Structure</b>Each dataset directory follows this structure:[Dataset Name]/<br>├── images/ # UAV-captured image frames<br>├── annotations_yolo/ # YOLO-format annotations<br>├── annotations_coco/ # COCO-style JSON annotations<br>├── README_[Dataset].md # Dataset-specific documentation<b>3. Annotation Formats</b>YOLO Format:<br><code>[class_id x_center y_center width height]</code> — normalized coordinates per object.COCO Format:<br><code>images</code>, <code>annotations</code>, and <code>categories</code> fields, compatible with standard tools like <code>pycocotools</code>, MMDetection, and Detectron2.<b>4. Use Cases</b>Small object detection in thermal and RGB drone imageryEcological monitoring and conservation AIMulti-class and fine-grained object recognitionEvaluation of domain adaptation and model generalization under real-world UAV scenarios<br>