Pith. sign in

Pixel-wise Agricultural Image Time Series Classification: Comparisons and a Deformable Prototype-based Approach

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it
abstract

Improvements in Earth observation by satellites allow for imagery of ever higher temporal and spatial resolution. Leveraging this data for agricultural monitoring is key for addressing environmental and economic challenges. Current methods for crop segmentation using temporal data either rely on annotated data or are heavily engineered to compensate the lack of supervision. In this paper, we present and compare datasets and methods for both supervised and unsupervised pixel-wise segmentation of satellite image time series (SITS). We also introduce an approach to add invariance to spectral deformations and temporal shifts to classical prototype-based methods such as K-means and Nearest Centroid Classifier (NCC). We study different levels of supervision and show this simple and highly interpretable method achieves the best performance in the low data regime and significantly improves the state of the art for unsupervised classification of agricultural time series on four recent SITS datasets.

fields

cs.CV 1

years

2026 1

verdicts

CONDITIONAL 1

representative citing papers

SwissCrop25: A National Multi-Year Benchmark for Operational Crop Mapping

cs.CV · 2026-08-10 · conditional · novelty 7.0

A seven-year Swiss crop mapping benchmark with a leave-one-year-out protocol reveals that a spatio-temporal transformer (TSViT) beats a convolutional temporal-attention model (U-TAE) by 12 points in macro-mIoU, while an Earth observation foundation model (Galileo) trails both.

citing papers explorer

Showing 1 of 1 citing paper.

  • SwissCrop25: A National Multi-Year Benchmark for Operational Crop Mapping cs.CV · 2026-08-10 · conditional · none · ref 48 · internal anchor

    A seven-year Swiss crop mapping benchmark with a leave-one-year-out protocol reveals that a spatio-temporal transformer (TSViT) beats a convolutional temporal-attention model (U-TAE) by 12 points in macro-mIoU, while an Earth observation foundation model (Galileo) trails both.