Fusing embeddings from four Earth models (AlphaEarth, Tessera, GeoCLIP, SatCLIP) outperforms the best single model on four of six tasks, with gains depending on task and location.
TESSERA: Temporal Embeddings of Surface Spectra for Earth Representation and Analysis
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
abstract
Satellite Earth-observation (EO) time series in the optical and microwave ranges of the electromagnetic spectrum are often irregular due to orbital patterns and cloud obstruction. Compositing addresses these issues but loses information with respect to vegetation phenology, which is critical for many downstream tasks. Instead, we present TESSERA, a pixel-wise foundation model for multi-modal (Sentinel-1/2) EO time series that learns robust, label-efficient embeddings. During model training, TESSERA uses Barlow Twins and sparse random temporal sampling to enforce invariance to the selection of valid observations. We employ two key regularizers: global shuffling to decorrelate spatial neighborhoods and mix-based regulation to improve invariance under extreme sparsity. We find that for diverse classification, segmentation, and regression tasks, TESSERA embeddings deliver state-of-the-art accuracy with high label efficiency, often requiring only a small task head and minimal computation. To democratize access, adhere to FAIR - principles, and simplify use, we release global, annual, 10m, pixel-wise int8 embeddings together with open weights/code and lightweight adaptation heads, thus providing practical tooling for large-scale retrieval and inference at planetary scale. All code and data are available at: https://github.com/ucam-eo/tessera.
citation-role summary
background 2
citation-polarity summary
years
2026 5verdicts
UNVERDICTED 5roles
background 2polarities
background 2representative citing papers
FLUXtrapolation is a benchmark for domain generalization in ecosystem flux upscaling using temporal, spatial, and temperature-based extrapolation scenarios, with pilot results showing model separation on tail and multi-scale metrics.
Agentic AI faces structural challenges in remote sensing due to geospatial data properties and workflow constraints, requiring EO-native agents built around structured state, tool-aware reasoning, and validity-aware evaluation.
SSDM decouples global geospatial embeddings into structural modulation and semantic injection pathways to improve accuracy and consistency in high-resolution remote sensing land cover mapping.
LIANet encodes multi-temporal Earth observation data into a coordinate-based neural field that supports label-only fine-tuning for downstream tasks without access to raw imagery.
citing papers explorer
-
Better Together: Evaluating the Complementarity of Earth Embedding Models
Fusing embeddings from four Earth models (AlphaEarth, Tessera, GeoCLIP, SatCLIP) outperforms the best single model on four of six tasks, with gains depending on task and location.
-
FLUXtrapolation: A benchmark on extrapolating ecosystem fluxes
FLUXtrapolation is a benchmark for domain generalization in ecosystem flux upscaling using temporal, spatial, and temperature-based extrapolation scenarios, with pilot results showing model separation on tail and multi-scale metrics.
-
Agentic AI for Remote Sensing: Technical Challenges and Research Directions
Agentic AI faces structural challenges in remote sensing due to geospatial data properties and workflow constraints, requiring EO-native agents built around structured state, tool-aware reasoning, and validity-aware evaluation.
-
Structure-Semantic Decoupled Modulation of Global Geospatial Embeddings for High-Resolution Remote Sensing Mapping
SSDM decouples global geospatial embeddings into structural modulation and semantic injection pathways to improve accuracy and consistency in high-resolution remote sensing land cover mapping.
-
Location Is All You Need: Continuous Spatiotemporal Neural Representations of Earth Observation Data
LIANet encodes multi-temporal Earth observation data into a coordinate-based neural field that supports label-only fine-tuning for downstream tasks without access to raw imagery.