Planetary Feature Fields are Scalable Earth Representations

2026-09-29

AI Summary

This study addresses the petabyte-scale redundancy and high access latency caused by the independent storage of satellite data by proposing Planetary Feature Fields, which model multi-source Earth observation data as spatiotemporally continuous functions. The method introduces a novel explicit-implicit hybrid neural field architecture that achieves a jointly compressed representation through factorized 3D grid decomposition, shared feature volumes, and a lightweight implicit decoder, while supporting incremental expansion to new timesteps and products without degrading existing outputs. Experiments demonstrate that this approach retains over 90% of downstream task performance at a 1800× compression ratio and reduces end-to-end access latency by an order of magnitude.

Abstract

Satellite observations, precomputed embeddings, and map products describe the same evolving Earth, yet are stored as independent, petabyte-scale data products. Their continued growth calls for compact representations of multiple products while preserving spatial and temporal detail. We introduce Planetary Feature Fields (PFFs), which exploit redundancy across data products by modeling them jointly as continuous functions of space and time at planetary scale. PFFs are spatially local explicit-implicit (hybrid) neural fields. Each field shares a factored feature volume---a decomposition of an explicit 3D grid with smaller factors---across products, while lightweight implicit decoders reconstruct individual products across multiple timesteps. PFFs reconstruct EO products over space and time more accurately than single-product fields at matched compression rates. At $1800\times$ compression relative to the uncompressed source data, reconstructed features retain approximately $90\%$ or more of the performance achieved with the original features on pixel-level segmentation, change detection, and patch-level classification tasks. PFFs can add new timesteps by extending their factored feature volumes and add new products by attaching new decoders, while leaving existing outputs unchanged. PFFs reduce end-to-end feature access latency by an order of magnitude relative to evaluated API and cloud-storage pipelines.

Problem

• Earth observation
• data compression
• neural fields
• scalable representation
• spatiotemporal modeling

Innovation

• Planetary Feature Fields
• hybrid neural fields
• factored feature volume
• Earth observation compression
• scalable representation