TY - DATA
T1 - CODaN: Common Objects Day and Night, an image classification dataset.
PY - 2023/11/29
AU - Attila Lengyel
AU - Sourav Garg
AU - Michael Milford
AU - Jan van Gemert
UR - 
DO - 10.4121/05246b12-f291-460c-be8a-f14a89d249e5.v1
KW - color invariants
KW - day-night
KW - computer vision
KW - image classification
KW - domain adaptation
N2 - <p>Common Objects Day and Night (CODaN) is an image classification dataset for zero-shot day-night domain adaptation / generalization.</p><p><br></p><p>The CODaN dataset consists of 15,500 224x224 colour images in 10 classes, with 1,550 images per class. There are 10,000 training images, 500 validation images, 2,500 daytime test images and 2,500 nighttime test images.</p><p><br></p><p>The dataset is collected from the excellent&nbsp;<a href="https://cocodataset.org/#home" target="_blank" style="color: rgb(31, 35, 40);">COCO</a>,&nbsp;<a href="http://imagenet.stanford.edu/" target="_blank" style="color: rgb(31, 35, 40);">ImageNet</a>&nbsp;and&nbsp;<a href="https://github.com/cs-chan/Exclusively-Dark-Image-Dataset" target="_blank" style="color: rgb(31, 35, 40);">ExDark</a>&nbsp;datasets. All images are filtered and cropped such that they have the same dimensions and are completely mutually exclusive, i.e. do not contain objects of different classes, nor do belong objects to multiple classes.</p>
ER -