Scientific publications
The following table provides overview of scientific contributions sourced from Edge-SpAIce developments.
Publication | Journal / Event | Authors | Date |
|---|---|---|---|
HLS4ML: A Flexible, Open-Source Platform for Deep Learning Acceleration on Reconfigurable Hardware | ACM Transactions on Reconfigurable Technology and Systems | Jan-Frederik Schulte, Benjamin Ramhorst, Chang Sun, Jovan Mitrevski, Nicolò Ghielmetti, Enrico Lupi, Dimitrios Danopoulos, Vladimir Loncar, Javier Duarte, David Burnette, Lauri Laatu, Stylianos Tzelepis, Konstantinos Axiotis, Quentin Berthet, Haoyan Wang, Paul White, Suleyman Demirsoy, Marco Colombo, Thea Aarrestad, Sioni Summers, Maurizio Pierini, Giuseppe Di Guglielmo, Jennifer Ngadiuba, Javier Campos, Ben Hawks, Abhijith Gandrakota, Farah Fahim, Nhan Tran, George Constantinides, Zhiqiang Que, Wayne Luk, Alexander Tapper, Duc Hoang, Noah Paladino, Philip Harris, Bo-Cheng Lai, Manuel Valentin, Ryan Forelli, Seda Ogrenci, Lino Gerlach, Rian Flynn, Mia Liu, Daniel Diaz, Elham Khoda, Melissa Quinnan, Russell Solares, Santosh Parajuli, Mark Neubauer, Christian Herwig, Ho Fung Tsoi, Dylan Rankin, Shih-Chieh Hsu, Scott Hauck | 2026.05.27 published |
Edge SpAIce: Deep Neural Network simplification pipeline for On-Board Data Reduction – PAPER | European Data Handling & Data Processing Conference for Space (EDHPC 2025) | François De Vieilleville,
Nicolas-Marcel Lemoine,
Pierre-Jean Coquard,
Pauline Audenino,
Sioni Paris Summers,
Boyan-Nikola Zafirov,
Simon Vellas | 2026.01.19 published |
Edge SpAIce: Deep Neural Network simplification pipeline for On-Board Data Reduction – SLIDES | European Data Handling & Data Processing Conference for Space (EDHPC 2025) | Nimesh Tahalooa | 2025.10.17 presented |
Real-time detection and monitoring of plastics pollution with AI onboard Earth Observation Satellites | 23rd International Workshop on Advanced Computing and Analysis Techniques in Physics Research (ACAT 2025) | Sioni Summers | 2025.09.11 presented |
Edge SpAIce: Enabling Onboard Data Compression With Machine Learning On FPGAs | International Parallel and Distributed Processing Symposium Workshops (IPDPSW 2025) | Noemi D’Abbondanza, Stylianos Tzelepis, Nicolo Ghielmetti, Ioannis Kakogeorgiou, Vanya Buchova, Konstantinos Karantzalos, Katerina Kikaki, Nicolas-Marcel Lemoine, Maurizio Pierini, Sioni Summers, Simon Vellas, Francois de Vieilleville, Boyan-Nikola Zafirov | 2025.06.07 presented,
2025.08.13 published |
Trialing Real-Time Global Marine Litter Monitoring With Edge-SpAIce Project (POSTER) | Living Planet Symposium (LPS 2025) – ESA | Dr. Andis Dembovskis,
Dr. François de Vieilleville,
Dr. Pauline Audenino,
Dr. Sioni Summers,
Dr. Kikaki Katerina,
Boyan-Nikola Zafirov | 2025.06.23 |
Enabling Onboard Data Compression with Machine Learning on FPGAs | SpacE FPGA Users Workshop (SEFUW 2025) | Nicolò Ghielmetti,
Dr Maurizio Pierini,
Ms Noemi D’Abbondanza,
Dr Sioni Summers,
Stylianos Tzelepis | 2025.03.25 |
Enabling Onboard Data Compression with Machine Learning on FPGAs | Applied Machine Learning Days (AMLD 2025) | S. Tzelepis, N. Ghielmetti, N. M. Lemoine, M. Pierini, S. Summers, F. De Vielleville | 2025.02.11 |
Edge SpAIce: Enabling On-Board Data Compression With Machine Learning On FPGAs | Fast Machine Learning for Science Conference 2024 | Nicolò Ghielmetti, Mr Stylianos Tzelepis, Nicolò Ghielmetti, Sioni Paris Summers,
Maurizio Pierini | 2024.10.16 |
Edge SpAIce: Enabling Onboard Data Compression With Machine Learning On FPGAs | International Workshop on On-Board Payload Data Compression (OBPDC 2024) | Tzelepis, Stylianos; Ghielmetti, Nicolò; Lemoine, Nicolas-Marcel; Pierini, Maurizio; Summers, Sioni; De Vielleville, François
| 2024.09.30 |
Other publicity
The following table provides overview of published material about Edge-SpAIce project in various conferences, exhibitions, workshops, tutorials and events, aimed at promoting the project publicity.
Publication | Event | Contents | Authors | Date |
|---|---|---|---|---|
Tutorial on hls4ml (videorecording + slides) | Fast Machine Learning for Science Conference 2026 | This tutorial covers the full hls4ml model conversion pipeline, from the training of small example models, model conversion into HLS code, and HLS synthesis to obtain FPGA resource estimates. Including model compression with tools such as QKeras v3 and HGQ, and the impact on the inference performance on FPGAs. | Chang Sun (California Institute of Technology);
Georgios Flengas (CERN);
Jan-Frederik Schulte (Purdue University);
Marius Köppel (ETH Zurich) | 2026.08.31 |
EdgeSpAIce contribution to EuroGEO workshop | EuroGEO – Data to Intelligence: co-creating strategies with EU Initiatives. Marine & Coastal Action Group Workshop Data-to-Intelligence. | Current high level summary about Edge-SpAIce, focusing on preserving environment | Pierini Maurizio; Enrico Chesta | 2026.07.06 |
Edge-SpAIce slides, focus on AI | Rendez-vous Mer & Données | Rencontre Marché IA & Maritime | Current high level summary about Edge-SpAIce, focusing on AI-tech | Andis Dembovskis | 2026.04.09 |
Tutorial on hls4ml | FPGA Developers’ Forum (FDF) | This tutorial provides a practical introduction to hls4ml for FPGA-based machine learning inference. Participants will go through the main hls4ml workflow using small neural network examples, learn how key configuration choices affect latency and resource usage, and explore basic optimization techniques such as quantization and pruning. The hands-on parts of the tutorial runs on a CERN-curated Jupyter notebook platform. | Georgios Flengas | 2026.05.29 |
Tutorial on hls4ml | Fast Machine Learning for Science Conference 2025 | This tutorial introduces and gives a hands-on demo on hls4ml, an open-source library for real-time deployment of neural networks on FPGAs. hls4ml allows a seamless conversion from high-level models (e.g., from Keras or PyTorch) to low-latency, low-power FPGA designs. The tutorial covers the design choices behind hls4ml, from deeply pipelined dataflow architectures to model quantization and pruning. The hands-on demo guides through experimenting with hls4ml’s Python API and concludes with a live demo of the model inference on a real FPGA. | Benjamin Ramhorst (ETH Zurich);
Fast Machine Learning Collaboration community | 2025.09.01 |
Edge-SpAIce event slides @LPS | Living Planet Symposium (LPS 2025) – ESA | All presentations combined from the Edge-SpAIce event @LPS’25 | Consortium members & event speakers | 2025.06.24 |
Edge-SpAIce poster @HaDEA/LPS | Living Planet Symposium (LPS 2025) – ESA | High level overview of Edge-SpAIce | Andis Dembovskis | 2025.06.23 |
Edge-SpAIce poster @GEO-Forum | GEO Global Forum 2025 | High-level overview of Edge-SpAIce activities | S. Tzelepis; N. Ghielmetti; M. Pierini; S. Summers; K. Kikaki; S. Vellas; I.Kakogeorgiou; N.M. Lemoine; F. De Vielleville; A. Dembovskis; P. Audenino; B. Zafirov | 2025.05.05 |
Edge-SpAIce slides @HaDEA | HaDEA workshop | Overview and progress of Edge-SpAIce project | Andis Dembovskis | 2025.02.21 |
Edge-SpAIce overview (slides, video recording) | CERN Knowledge Transfer Seminar | Overview and progress of Edge-SpAIce | Nick Ziogas, Sioni Summers, Konstantinos Karantzalos, Francois de Vieilleville, Joseph Thuilier | 2024.10.29 |
Public resources
The following table provides overview of published material about Edge-SpAIce project in various conferences, exhibitions, workshops, tutorials and events, aimed at promoting the project publicity.
Resource | Notes |
|---|---|
MADOS – Detecting Marine Pollutants and Sea Surface Features with Deep Learning in Sentinel-2 Imagery | This is a package of the DNN-training dataset, DNN architecture and DNN itself for anyone interested in reproducing AI training for marine plastic detection in Sentinel-2 images. This pre-existing resource is updated with learnings from Edge-SpAIce. |
MARIDA – Maritime Debris Archive | Marine Debris Archive (MARIDA) is a marine debris-oriented dataset on Sentinel-2 satellite images. It also includes various sea features that co-exist. MARIDA is primarily focused on the weakly supervised pixel-level semantic segmentation task. |
HLS4ML code, installation guide and tutorials | Here can be found open-source code of HLS4ML for deploying it on NanoXplore NG-Ultra SoC-FPGA. While HLS4ML existed before he Edge-SpAIce, it was during this Horizon Europe project that the HLS4ML was augmented with feasibility to deploy on the sovereign European SoC-FPGA. |
Follow us
To enable anyone outside project to follow our consortium’s activities we have created @edgespaice page on LinkedIn – feel free to follow us. We post about events we are planning to attend, any insights from conferences and consortium meetings.