: It provides three levels of spatial annotations: binary vessel masks, general vein-artery masks, and fine-grained vein-artery masks.
The search result for typically refers to the Handheld Device-Based Fundus Video Dataset (RVD) , which is the official repository for the paper titled " Leveraging Frame Affinity for Retinal Vessel Segmentation in Fundus Videos " published at CVPR 2024. rvd.zip
: 635 smartphone-based videos from four different clinics, featuring patients aged 50–75. : It provides three levels of spatial annotations:
: The RVD.zip file typically contains two parts, each further divided into train and test sets containing RAW data, flow information, and sRGB frames. Other Academic Contexts for "RVD" : The RVD
If your request is not related to image processing, "RVD" is used as an acronym in several other academic fields:
This dataset is specifically designed for tasks using videos recorded on mobile devices. Key Details of the Dataset