TerraCodec: Compressing Earth Observations

Julen Costa Watanabe, Isabelle Wittmann, Benedikt Blumenstiel, Konrad Schindler

Keywords: Earth Observation, Compression, Pretraining

TL;DR: We present TerraCodec, a family of learned models that compress satellite imagery by exploiting multispectral and temporal structure, achieving higher efficiency than standard image or video codecs.

Abstract: Earth observation (EO) satellites produce massive streams of multispectral image time series, posing pressing challenges for storage and transmission. Yet, learned EO compression remains fragmented and lacks publicly available, large-scale pretrained codecs. Moreover, prior work has largely focused on image compression, leaving temporal redundancy and EO video codecs underexplored. To address these gaps, we introduce TerraCodec (TEC), a family of learned codecs pretrained on Sentinel-2 EO data. TEC includes efficient multispectral image variants and a Temporal Transformer model (TEC-TT) that leverages dependencies across time. To overcome the fixed-rate setting of today's neural codecs, we present Latent Repacking, a novel method for training flexible-rate transformer models that operate on varying rate-distortion settings. TerraCodec outperforms classical codecs, achieving 3-10x stronger compression at equivalent image quality. Beyond compression, TEC-TT enables zero-shot cloud inpainting, surpassing state-of-the-art methods on the AllClear benchmark. Our results establish EO-trained neural codecs and temporal compression as a promising direction for Earth observation. Code and model weights will be released under a permissive license.

Primary Area: applications to computer vision, audio, language, and other modalities

Submission Number: 11569