RT-Trees: Thermal training images

This is the RT-Trees dataset proposed and used in the paper titled, "Shadowsense: Unsupervised Domain Adaptation and Feature Fusion for Shadow-Agnostic Tree Crown Detection From RGB-Thermal Drone Imagery", published at the IEEE/CVF WACV 2024 conference. Due to the size of the dataset and Zenodo's 50GB limit, the dataset is partitioned into two separate uploads. This upload contains the unlabelled thermal images used for unsupervised training

The first upload includes evaluation splits (test & val), along with the labelled subset of RGB training images used for a supervised training experiment, and the much larger set of unlabelled RGB images used for fully-unsupervised training.