I am a fourth year ETH AI Center Doctoral fellow, supervised by Prof. Konrad Schindler (ETH, PRS) and Prof. Jan Dirk Wegner (UZH, Ecovision). My research lies at the frontier of computer vision and remote sensing to solve scientific questions in the environmental sciences. In particular, my current work focuses on high-resolution (10m) global mapping of above-ground biomass.

I am also the co-founder and an active member of the AI + Environment Summit, a non-profit that organizes the eponymous event.

I did my MSc at ETH Zurich, and my undergrad at École Polytechnique (Paris).

From Machine Learning to large-scale EO products: Good Practices and Guidelines for making Maps (In preparation)

Ghjulia Sialelli et al.

🔗 Project website

🔗 ECCV workshop paper

The past decade has seen a rapid expansion in the production of large-scale geospatial products derived from Earth observation (EO) data using machine learning (ML). We present an end-to-end account of good practices, challenges, and common pitfalls that span the entire pipeline from raw satellite data to a published, validated map product.

Multi-Modal, Multi-Temporal Geospatial Foundation Models Improve Global Above-Ground Biomass Estimation
(Under review)

Ghjulia Sialelli, Linus Scheibenreif, Jan D. Wegner, Konrad Schindler

🔗 Preprint

🔗 Project website

🔗 NeurIPS workshop paper

We benchmark 13 GFMs for AGB estimation and compare them against a fully supervised baseline. Most of the GFMs substantially underperform, whether frozen or fine-tuned; only TESSERA and AlphaEarth Foundations features yield a better performance.

Gaussian Process Regression for Bias Correction of Satellite-based Biomass Estimates (Under review)

Ghjulia Sialelli, Torben Peters, Linus Scheibenreif, Jan D. Wegner, Konrad Schindler

🔗 Preprint

🔗 Project website

To address the spatially structured residuals and biases typical of large-scale AGB products, we develop a post-hoc calibration scheme using Gaussian Process Regression.

AGBD: A Global-scale Biomass Dataset (ISPRS GSW 2025 Oral)
Ghjulia Sialelli, Torben Peters, Jan D. Wegner, Konrad Schindler

🔗 PDF

🔗 Project website

We introduce a ML-ready dataset and benchmark models, for dense high-resolution global biomass estimation.

Disaster Displacement in Asia and the Pacific (2022)

Direction and project management: Alexandra Bilak, Bina Desai, and Christelle Cazabat. Project Coordination: Vicente Anzellini and Fanny Teppe. Drafting and Research: Vicente Anzellini, Christelle Cazabat, Pablo Cortés Ferrández, Ricardo Fal-Dutra Santos, Vincent Fung, Kathryn Giffin, Thannaletchimy Housset, Alesia O’Connor, Fanny Teppe, and Louisa Yasukawa. Data and Analysis: Sylvain Ponserre, Ghjulia Sialelli and Fanny Teppe. Design, layout, maps and graphs: Vivcie Bendo, Stéphane Kluser (Komplo), Emiliano Pérez, Sylvain Ponserre. Editor: Melanie Kelleher

🔗 Report

This report presents the disaster displacement trends in the region during 2010−2021 and provides insights into its social and economic impacts.