This dataset includes Thermal-Visual images with the same aligned geometry taking by FLIR-E60 camera. Images are of size (320 x 240) pixel resolution with 0.05◦C thermal sensitivity and −20◦C to 650◦C. Thermal images were extracted in their raw format and logged in 16-bit float per-pixel in one channel. The acquisition was made in different scenes and different environments (indoor and outdoor, during winter and summer and with static and moving objects). Thermal and RGB images were manually extracted and annotated with a total of 570 pairs in the original paper. Due to European regulations about data and privacy protection, Images shows identified faces were excluded from this published version. This published benchmark has 404 pair of images, divided into 280 training samples, 78 validation validation samples and 46 testing samples. The file 'ULB17-VT.pkl' is saved in Pkl format, using python 3.5 and cPickle.HIGHEST_PROTOCOL and arranged in the following way: [Train_RGBx, Train_HRx, Train_LRx], [Valid_RGBx, Valid_HRx, Valid_LRx], [Test_RGBx, Test_HRx, Test_LRx] X_RGBx: Visual RGB image X_HRx: High-Resolution Thermal image X_LRx: Low-Resolution 'Gaussian Pyramid' down-sampled Thermal image This dataset is first presented in the article below, Please cite this paper if the dataset is used in your publication. @article{almasri2018multimodal, title={Multimodal Sensor Fusion In Single Thermal image Super-Resolution}, author={Almasri, Feras and Debeir, Olivier}, journal={arXiv preprint arXiv:1812.09276}, year={2018} }

{"references": ["Almasri, Feras et al. (2018). Multimodal Sensor Fusion In Single Thermal image Super-Resolution to arXiv:1812.09276"]}