Panoptic Segmentation of Satellite Image Time Series with Convolutional Temporal Attention Networks

TL;DR

This paper introduces a novel end-to-end method for panoptic segmentation of satellite image time series, leveraging temporal self-attention to capture complex crop phenology patterns, and provides the first open-access dataset for this task.

Contribution

It presents the first single-stage panoptic segmentation approach for satellite image time series and introduces PASTIS, the first open-access dataset with panoptic annotations.

Findings

Our encoder outperforms competing architectures in semantic segmentation.

The method achieves state-of-the-art results in panoptic segmentation of SITS.

Implementation and dataset are publicly available.

Abstract

Unprecedented access to multi-temporal satellite imagery has opened new perspectives for a variety of Earth observation tasks. Among them, pixel-precise panoptic segmentation of agricultural parcels has major economic and environmental implications. While researchers have explored this problem for single images, we argue that the complex temporal patterns of crop phenology are better addressed with temporal sequences of images. In this paper, we present the first end-to-end, single-stage method for panoptic segmentation of Satellite Image Time Series (SITS). This module can be combined with our novel image sequence encoding network which relies on temporal self-attention to extract rich and adaptive multi-scale spatio-temporal features. We also introduce PASTIS, the first open-access SITS dataset with panoptic annotations. We demonstrate the superiority of our encoder for semantic segmentation against multiple competing architectures, and set up the first state-of-the-art of panoptic segmentation of SITS. Our implementation and PASTIS are publicly available.