Self-supervised contrastive learning on agricultural images

Ronja Güldenring, Lazaros Nalpantidis

Abstract: Highlights•Generation of agricultural pre-trained weights with self-supervised contrastive learning SwAV for three different agricultural datasets.•The learned pre-trained weights are transferred to classification as well as segmentation downstream tasks.•We show that domain-specific pre-trained weights increase the label efficiency of classification tasks significantly.