Satellite image deep learning – Details, episodes & analysis

In this episode I sat down with Mikolaj (Miko) Czerkawski from Asterisk Labs to explore BetaEarth, an experimental open-source emulator trained on AlphaEarth Foundations' public embedding archive. AEF — released by Google and Google DeepMind as a global 10 m embedding product derived from a wide range of Earth-observation modalities — is what makes BetaEarth possible: its openness lets the community build lightweight independent emulators that approximate AEF's pixelwise outputs from standard Sentinel inputs, and use them to probe how much of a model's behaviour is captured in its public embeddings. Miko walks through BetaEarth's design — compact architectures based on SegFormer-B2 with separate per-modality encoders, and a shared DINOv3 backbone over 3-band spectral primitives — and the surprising finding that reasonably strong approximations can be achieved even from simple RGB inputs.We then dive into a live demo: generating BetaEarth embeddings for arbitrary regions and time ranges using Sentinel-1, Sentinel-2, and COP-DEM data. Along the way, we cover practical considerations such as cloud contamination, modality trade-offs, tiling artefacts, and strategies for merging multi-temporal signals. Finally, we discuss what this complementary tooling enables for the geospatial ML community — embeddings as pretraining or regularisation signals, lightweight local inference alongside AEF's global annual rasters, and what the combination of large proprietary archives and open emulator-style tools could unlock next.

* 📺 Video of this conversation & demo on YouTube

* 🖥️ BetaEarth Github page

* 🖥️ BetaEarth demo on Huggingface

Bio: Miko is a researcher specialising in AI, computer vision, signal processing and Earth observation. Before co-founding Asterisk Labs he was a postdoctoral research fellow at the European Space Agency. His research interests include data-centric analyses of large-scale Earth observation data, dataset curation, generative modelling, and restoration tasks for satellite imagery. He is a co-founder of the Major TOM community project, a platform for collaborating and reusing Earth observation datasets designed specifically for AI pipelines. He received the B.Eng. degree in electronic and electrical engineering in 2019 from the University of Strathclyde in Glasgow, United Kingdom, and the Ph.D. degree in 2023 at the same institution, specialising in applications of computer vision to Earth observation data.

In this episode I sat down with Kentaro Wada, a computer vision engineer at Mujin and creator of LabelMe, to explore the evolution of image annotation workflows. We discuss how his need to label data for a robotics challenge led to building one of the most widely used open-source annotation tools, and how it has evolved alongside the shift from traditional computer vision to deep learning. Kentaro explains the impact of foundation models like Segment Anything (SAM), and how annotation is rapidly moving toward a prompt-and-verify paradigm where models do the heavy lifting and humans focus on quality control. We also dive into his recent work integrating SAM into LabelMe, the challenges of applying these models to satellite imagery, and why approaches like bounding-box prompting outperform text in that domain. Finally, we cover new support for large, multi-channel geospatial data, practical deployment considerations, and what this means for scaling annotation in real-world machine learning systems. Note that a recording of this conversation, along with a demonstration of geospatial annotation using LabelMe, is available on YouTube via the links below:

* 🖥️ LabelMe website

* 🖥️ Kentaro’s personal website

* 📺 Video of this conversation on YouTube

* 📺 Demo video on YouTube

Bio: Kentaro Wada was born in Japan in 1994. He received his B.Sc. (2016) and M.Sc. (2018) from Mechanical Engineering and Computer Science Department in The University of Tokyo (UTokyo). In his research at UTokyo, he was working on learning-based scene understanding for robotic manipulation at JSK Laboratory supervised by Prof. Masayuki Inaba and Prof. Kei Okada. He received his PhD in 2022, at Dyson Robotics Laboratory in Imperial College London supervised by Prof. Andrew Davison. During his PhD, he worked on object-level semantic scene understanding, a general scene representation useful for robotic manipulation, and showed several novel capabilities of robots. He joined Mujin, Inc. in 2022 as a computer vision engineer, and is working on advancing robots' capabilities in the real-world environment.

In this episode I caught up with Adam Stewart, creator of TorchGeo, to hear all the latest updates related to this pivotal piece of geospatial AI software. We discuss TorchGeo’s strong adoption in the geospatial ML community and the upcoming 1.0 release, which will introduce long-awaited time series support. Adam shares insights from a recent software literature review covering available geospatial data handling tools, sampling strategies, and the broader machine learning ecosystem. He also talks about the newly formed Technical Steering Committee, outlining its role in guiding the project’s direction. Other topics include upcoming breaking changes to geospatial datasets and samplers, how TorchGeo integrates with other libraries and tools, the project’s growing community, the role of foundation models in handling diverse geospatial products, the promise of zero-shot learning for effortless data labelling, and why no single model can dominate across all domains.

* 👤 Adam on LinkedIn

* 🖥️ TorchGeo

* 📺 Video of this conversation on YouTube

Bio: Adam J. Stewart's research interests lie at the intersection of machine learning and Earth science, especially remote sensing. He is the creator and lead developer of the popular TorchGeo library, a PyTorch domain library for working with geospatial data and satellite imagery. His current research focuses on building foundation models for multispectral imagery. He received his B.S. from the Department of Earth and Atmospheric Sciences at Cornell University and his Ph.D. from the Department of Computer Science at the University of Illinois Urbana-Champaign. He currently works as a postdoctoral researcher at the Technical University of Munich under the guidance of Prof. Xiaoxiang Zhu.

In this episode I caught up with Tobias Augspurger to explore the Map Your Grid initiative at Open Energy Transition, an ambitious project funded by Breakthrough Energy to build a digital twin of the global electrical grid. While AI and machine learning are being used to detect substations, pylons, and transmission lines in satellite imagery, Toby explains why these approaches alone can’t deliver a complete, accurate map. We discussed the false positives, missing connections, and contextual details that challenge automated models, and how human validation and open-source mapping remain essential to producing reliable, global-scale infrastructure data.

* 👤 Toby on LinkedIn

* 🖥️ mapyourgrid.org

* 📺 MapYourGrid YouTube Channel

* 📺 Video of this conversation on YouTube

Bio: Tobias Augspurger is a climate technology innovator and open-source advocate. At Open Energy Transition, he is accelerating the global energy transition by standardising electrical grid data within OpenStreetMap as part of the MapYourGrid initiative. With a PhD in atmospheric sciences and a background in aerospace engineering, Tobias combines technical expertise in remote sensing with inclusive collaboration. In his spare time, he works on OpenSustain.tech and ClimateTriage.com, connecting and promoting open projects to combat climate change and biodiversity loss

In this episode, I catch up with Federico Bessi to dive into a fascinating end-to-end project on the automatic detection of photovoltaic (PV) solar plants using satellite imagery and deep learning. Federico walks us through how he built a complete pipeline—from sourcing and preprocessing data using the Brazil Data Cube, to annotating solar farms in QGIS, training models in PyTorch, and finally deploying a web app on AWS to visualise the predictions.

This is interesting because solar energy infrastructure is expanding rapidly, yet tracking it globally remains a major challenge. This project demonstrates how open data and modern ML tools can be combined to monitor solar installations at scale—automatically and remotely. It's a compelling example of applied geospatial AI in action.

This video is ideal for remote sensing practitioners, machine learning engineers, and anyone interested in environmental monitoring, Earth observation, or building practical AI systems for real-world deployment.

* 🖥️ Project code on Github

* 👤 Federico on Linkedin

* 📺 Video of this conversation on YouTube

* 📺 Project demo on YouTube

Bio: Federico Bessi is a Software Engineer specializing in Machine Learning, with an international background in the software, computer vision, and biometrics industries. He spent over a decade working in biometric identification for global tech companies, contributing to national ID systems across more than seven countries. In these roles, he developed software, led engineering teams, and oversaw large-scale system operations. Building on this foundation, Federico has deepened his work in machine learning and deep learning, applying it to business intelligence, user satisfaction modeling, and geospatial analysis using satellite imagery. He also became a contributor with the open-source TorchGeo project.

In this conversation, I caught up with Shahab Jozdani to learn about Chat2Geo, a web-based application designed to simplify remote-sensing-based geospatial analysis through an intuitive, chatbot-style interface.

Large language models, such as ChatGPT, are reshaping the way users interact with complex datasets, and it’s inspiring to see innovators like Shahab leverage this technology to democratise geospatial analytics. Note that we also recorded a demonstration video of Chat2Geo, which is linked below:

* 🖥️ Chat2Geo on Github

* 👤 Shahab on LinkedIn

* 📺 Video of this conversation on YouTube

* 📺 Demo of Chat2Geo on YouTube

Bio: Data Scientist and Geomatics Engineer with over 10 years of experience in academia and industry, specialising in AI, computer vision, data science, software development, and building new solutions. Founder of GeoRetina, a Canadian company that developed and open-sourced Chat2Geo, an AI-powered platform providing real-time geospatial insights via conversational interfaces

In this episode, I caught up with Nick Wright to discuss OmniCloudMask—a Python library for state-of-the-art cloud and cloud shadow masking in satellite imagery.

Accurate cloud masking is crucial for reliable downstream analytics, yet creating models that generalise well across different sensors, resolutions, and atmospheric conditions remains a significant challenge.

OmniCloudMask addresses this through a novel image preprocessing pipeline and clever augmentation strategies that vary the image resolution presented to the model. Model generalisation is a key concern for practitioners in our field, and I found this conversation both insightful and practical—I hope you do too.

* 📃 Paper

* 🖥️ Code

* 📺 Video of this conversation on YouTube

* 👤 Nick on LinkedIn

Bio: Nick Wright is a Senior Research Scientist at the Western Australian Department of Primary Industries and Regional Development. He is also pursuing a PhD at the University of Western Australia, focusing on deep learning applications for environmental remote sensing, specifically in cloud and water detection and sensor-agnostic models.

In this episode, I caught up with James Doherty and Donal Hill, co-founders of Planetixx (formerly Plastic-i), a company using satellite imagery and AI to monitor ocean debris. Their platform not only detects plastic and other debris, but also predicts its origins and trajectory, enabling more effective interventions. Beyond plastic, they’ve expanded into monitoring algal blooms, a growing environmental concern.

The conversation covers the technical and practical challenges of building AI models that work at a global scale, as well as their newly launched platform. A live demo of the platform is available as a separate video, linked below

* 🖥️ Planetixx website

* 👤 James LinkedIn

* 👤 Donal LinkedIn

* 📺 Video of this conversation on YouTube

* 📺 Platform demo on YouTube

Bio: Dr. James Doherty is CEO and Co-Founder of Earthshot-nominated enterprise Planetixx, where he drives environmental innovation in tackling marine plastic pollution and promoting ocean health. His unique expertise spans astronomy, data science, and law, combining scientific rigour with legal acumen. James holds a PhD in Astronomy, law degrees from the Universities of Cambridge and Oxford, and is a Science to Data Science (S2DS) Fellow. His professional background includes practising as a commercial lawyer at Eversheds Sutherland before applying his diverse skill set to environmental entrepreneurship.

Bio: Dr. Donal Hill is Chief Technical Officer and Co-Founder of Planetixx, where he leads technology development initiatives in satellite data and artificial intelligence applications. His expertise spans particle physics, data science, and AI mplementation across research and industry. Donal holds a PhD in Particle Physics from the University of Oxford and spent ten years at CERN's Large Hadron Collider. His distinguished career includes serving as a Marie Curie Fellow at École Polytechnique Fédérale de Lausanne (EPFL) and holding senior data scientist positions at UEFA and the Swiss Data Science Center, where he facilitates AI adoption for industry partners.

In this episode, I caught up with Kai Jeggle to discuss his experience pursuing a PhD at the intersection of machine learning and remote sensing. The conversation covers Kai's work on IceCloudNet, a deep learning model that reconstructs 3D cloud structures from 2D imagery with sparse depth measurements. Data fusion and sparse machine learning are fascinating topics. I learned a lot from this conversation, and I hope you do to.

* 👤 Kai on LinkedIn

* 📃 IceCloudNet paper

* 🖥️ Code

* 💾 Dataset

* 📺 Video of this conversation on YouTube

Bio: Kai is passionate about leveraging machine learning to tackle climate change. His research lies at the intersection of ML, remote sensing, and climate science. He studied industrial engineering and computer science before completing his PhD in Atmospheric Physics at ETH Zurich under Prof. Ulrike Lohmann, with visiting research stays at UV Valencia and the ESA Phi Lab. He also worked as a software engineer at the Stockholm-based MLOps startup LogicalClocks. Kai is a core team member and former vice-chair of Climate Change AI, a global non-profit that catalyses impactful work at the intersection of climate change and machine learning. In his next role, he will join the meteo data team at Dexter Energy in Amsterdam, working to improve renewable energy yield forecasts.

In this episode, I caught up with Daniele Rege Cambrin, the organiser of the SMAC earthquake detection challenge, and Giorgio Morales, its winner. The challenge invited participants to leverage Sentinel 1 satellite imagery to identify earthquake-affected areas and measure the strength of events, while promoting scalable and resource-efficient solutions.

Giorgio shared his innovative approach that secured first place, and we explored the effort behind designing and solving such a meaningful challenge. This conversation provides valuable insights into developing effective solutions and showcases the potential of satellite data in earthquake monitoring.

* 📺 Video of this conversation on YouTube

* 🖥️ SMAC website

* 🖥️ Website of Giorgio

Bio: Giorgio is a PhD candidate (ABD) in computer science at Montana State University and a current member of the Numerical Intelligent Systems Laboratory (NISL). He holds a BS in mechatronic engineering from the National University of Engineering, Peru, and an MS in computer science from Montana State University, USA. His research interests are Deep Learning, Explainable Machine Learning, Computer Vision, and Precision Agriculture.