14 Open Agriculture Computer Vision Datasets Mapped by Task and Sensor Type

A field guide catalogs 14 open agriculture computer vision datasets, organized by task such as crop and weed segmentation, orchard phenotyping, and plant disease classification. Each entry lists its media type, label fields, and a direct Hugging Face link, all mirrored in FiftyOne format on the Voxel51 organization. The collection spans RGB, RGB-D, multispectral, satellite, 3D point cloud, and video data, reflecting the challenges that field conditions pose for general-purpose vision models. The guide also flags licensing differences, noting that some datasets are non-commercial or inherit varying national licenses, so users should check terms before commercial training. The datasets are presented as groundwork for physical AI applications in farming.
This is an AI-generated summary. ShortSingh links to the original source for the complete article.

Discussion (0)
Log in to join the discussion and vote.
Log in