ThermalMix

ThermalMix is a multi-view, object-centric dataset of well-aligned RGB and thermal images captured from six common objects. It consists of three forward-facing scenes (face, hand, panel) and three 360-degree scenes (lion, pan, laptop), and includes about 360 images in total. Cross-modality calibration was employed prior to data capturing, leading to high-quality alignments between RGB and thermal images.

More information about ThermalMix can be found in our paper:

Mert Özer, Maximilian Weiherer, Martin Hundhausen, Bernhard Egger. Exploring Multi-modal Neural Scene Representations With Applications on Thermal Imaging, arXiv, 2024.