Time2Agri: Temporal Pretext Tasks for Agricultural Monitoring
This addresses label-efficient agricultural monitoring for remote sensing applications, representing a domain-specific incremental improvement.
The paper tackled the problem that existing self-supervised learning pretext tasks for remote sensing overlook the temporal characteristics of agricultural landscapes. They proposed three agriculture-specific temporal pretext tasks, achieving 69.6% IoU on crop mapping and reducing yield prediction error to 30.7% MAPE, outperforming baselines.
Self Supervised Learning(SSL) has emerged as a prominent paradigm for label-efficient learning, and has been widely utilized by remote sensing foundation models(RSFMs). Recent RSFMs including SatMAE, DoFA, primarily rely on masked autoencoding(MAE), contrastive learning or some combination of them. However, these pretext tasks often overlook the unique temporal characteristics of agricultural landscape, namely nature's cycle. Motivated by this gap, we propose three novel agriculture-specific pretext tasks, namely Time-Difference Prediction(TD), Temporal Frequency Prediction(FP), and Future-Frame Prediction(FF). Comprehensive evaluation on SICKLE dataset shows FF achieves 69.6% IoU on crop mapping and FP reduces yield prediction error to 30.7% MAPE, outperforming all baselines, and TD remains competitive on most tasks. Further, we also scale FF to the national scale of India, achieving 54.2% IoU outperforming all baselines on field boundary delineation on FTW India dataset.