LLVIP stands for A Visible-infrared Paired Dataset for Low-light Vision. It is a dataset of visible-infrared pairs for low-light vision, including 30,976 images (15,488 pairs), 24 night scenes, and 2 day scenes. This dataset can be used for research on computer vision tasks such as image conversion (visible light to infrared, infrared to visible light), image fusion, low-light pedestrian detection, and infrared pedestrian detection.