Search
Buy your pass
OnlineDiscovery - GeoAIWorkshop Badge available

Hands-on TESSERA: Time-series embeddings for geospatial analysis

  • Register

     

    * Register (or log in) to the Neural Network to add this session to your agenda or watch the replay

    • Date
    2 February 2026
    Timeframe
    14:00 - 15:30 CET Geneva
    Duration
    90 minutes
    • Days
      Hours
      Min
      Sec

    Satellite remote sensing enables a wide range of downstream applications, including habitat mapping, carbon accounting, and strategies for conservation and sustainable land use. However, satellite time series are voluminous and often corrupted, making them challenging to use: the scientific community’s ability to extract actionable insights is often constrained by the scarcity of labelled training datasets and the computational burden of processing temporal data. 

    The presentation will introduce TESSERA (Time-series Embeddings of Surface Spectra for Earth Representation and Analysis), an open foundation model that preserves spectral-temporal signals in 128-dimensional latent representations at 10-meter resolution globally. The model uses self-supervised learning to summarise petabytes of Earth observation data. TESSERA is shown to be label-efficient and closely matches or outperforms state-of-the-art alternatives. By preserving temporal phenological signals that are typically lost in conventional approaches, TESSERA enables new insights into ecosystem dynamics, agricultural food systems, and environmental change detection. Moreover, the open-source implementation supports reproducibility and extensibility, while the privacy-preserving design allows researchers to maintain data sovereignty. To current knowledge, TESSERA is unprecedented in its ease of use, scale, and accuracy: no other foundation model provides analysis-ready outputs, is open, and delivers global, annual coverage at 10m resolution using only spectral-temporal features at pixel level. 

     

    Session Objectives:

    By the end of this session, participants will be able to:

    • Describe the challenges of using satellite time-series data for environmental analysis.
    • Explain the principles of self-supervised learning applied in the TESSERA model.
    • Compare TESSERA’s spectral-temporal embeddings with conventional approaches in terms of label efficiency and performance.
    • Apply TESSERA embeddings to crop identification and canopy height modeling tasks using provided datasets.
    • Evaluate the suitability of TESSERA for specific geospatial tasks based on its design and outputs.

     

    Recommended Mastery Level / Prerequisites:

    • Basic understanding of remote sensing and Earth observation data.
    • Familiarity with machine learning concepts, especially embeddings and neural networks.
    • Experience with Python and Jupyter Notebooks.
    Share this session with your network
    • Session starts in
      Days
      Hours
      Min
      Sec
    Add to Calendar
    Discover more from programme stream
    Discovery - GeoAI
    GeoAI in Earth Engine: AlphaEarth Foundations and the Satellite Embedding...

    GeoAI in Earth Engine: AlphaEarth Foundations and the Satellite Embedding...

    14 January 2026 - 14:30 to 16:30

    Introduction to TESSERA: Time-series embeddings for geospatial analysis

    Introduction to TESSERA: Time-series embeddings for geospatial analysis

    22 January 2026 - 14:00 to 15:00

    Hands-on TESSERA: Time-series embeddings for geospatial analysis

    Hands-on TESSERA: Time-series embeddings for geospatial analysis

    2 February 2026 - 14:00 to 15:30

    Embedding workflows for Earth Observation tasks

    Embedding workflows for Earth Observation tasks

    25 February 2026 - 14:00 to 15:30

    Introduction to Terratorch

    Introduction to Terratorch

    19 March 2026 - 14:00 to 15:30

    Related stories

    Are you sure you want to remove this speaker?