Del Prete, Roberto und Daga, Gabriele und Leonard, Cedric und Channing, Georgia und Longépé, Nicolas (2026) Maya4: a Novel Large-Scale Multi-Level SAR Representation Dataset for Machine Learning. In: 16th European Conference on Synthetic Aperture Radar, EUSAR 2024. 16th European Conference on Synthetic Aperture Radar (EUSAR), 2026-06-08 - 2026-06-11, Baden-Baden, Germany. doi: 10.30420/456729113.
|
PDF
1MB |
Offizielle URL: https://ieeexplore.ieee.org/document/11677538
Kurzfassung
Maya4 is a cloud-native Sentinel-1 Stripmap SAR dataset that exposes four aligned processing states: raw, rc, rcmc, and focused az. Unlike existing public resources limited to focused imagery, Maya4 exposes the signal chain from Level-0 echoes to focused products, enabling operator-level learning and evaluation. The release con- tains about 2 TB of data with chunked/sharded storage and co-located metadata for scalable access. Maya4 sup- ports three broad classes of signal-domain use cases: compression, learned focusing, and cross-level representa- tion learning, together with benchmarking against conventional processors. The dataset is publicly available at https://huggingface.co/buckets/ESA-philab/Maya4.
| elib-URL des Eintrags: | https://elib.dlr.de/227503/ | ||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Dokumentart: | Konferenzbeitrag (Vortrag) | ||||||||||||||||||||||||
| Titel: | Maya4: a Novel Large-Scale Multi-Level SAR Representation Dataset for Machine Learning | ||||||||||||||||||||||||
| Autoren: |
| ||||||||||||||||||||||||
| Datum: | 10 Juni 2026 | ||||||||||||||||||||||||
| Erschienen in: | 16th European Conference on Synthetic Aperture Radar, EUSAR 2024 | ||||||||||||||||||||||||
| Referierte Publikation: | Ja | ||||||||||||||||||||||||
| Open Access: | Ja | ||||||||||||||||||||||||
| Gold Open Access: | Nein | ||||||||||||||||||||||||
| In SCOPUS: | Nein | ||||||||||||||||||||||||
| In ISI Web of Science: | Nein | ||||||||||||||||||||||||
| DOI: | 10.30420/456729113 | ||||||||||||||||||||||||
| Status: | veröffentlicht | ||||||||||||||||||||||||
| Stichwörter: | Machine Learning, Synthetic Aperture Radar, Representation Learning, TB Data, Synthetic Aperture Radar Datasets, Intermediate Levels, Raw Signal, Latent Space, Synthetic Aperture, Earth Observation, Range Resolution, Self-supervised Learning, Residual Phase, Coherent Imaging, Matched Filter, Range Compression, Slant Range, Azimuth Resolution, High Range Resolution, Single Look Complex, Inverse Discrete Fourier Transform, Chirp Rate | ||||||||||||||||||||||||
| Veranstaltungstitel: | 16th European Conference on Synthetic Aperture Radar (EUSAR) | ||||||||||||||||||||||||
| Veranstaltungsort: | Baden-Baden, Germany | ||||||||||||||||||||||||
| Veranstaltungsart: | internationale Konferenz | ||||||||||||||||||||||||
| Veranstaltungsbeginn: | 8 Juni 2026 | ||||||||||||||||||||||||
| Veranstaltungsende: | 11 Juni 2026 | ||||||||||||||||||||||||
| Veranstalter : | VDE | ||||||||||||||||||||||||
| HGF - Forschungsbereich: | Luftfahrt, Raumfahrt und Verkehr | ||||||||||||||||||||||||
| HGF - Programm: | Raumfahrt | ||||||||||||||||||||||||
| HGF - Programmthema: | Erdbeobachtung | ||||||||||||||||||||||||
| DLR - Schwerpunkt: | Raumfahrt | ||||||||||||||||||||||||
| DLR - Forschungsgebiet: | R EO - Erdbeobachtung | ||||||||||||||||||||||||
| DLR - Teilgebiet (Projekt, Vorhaben): | R - Künstliche Intelligenz, R - AI4SAR | ||||||||||||||||||||||||
| Standort: | Oberpfaffenhofen | ||||||||||||||||||||||||
| Institute & Einrichtungen: | Institut für Methodik der Fernerkundung > EO Data Science | ||||||||||||||||||||||||
| Hinterlegt von: | Leonard, Cedric | ||||||||||||||||||||||||
| Hinterlegt am: | 02 Okt 2026 10:47 | ||||||||||||||||||||||||
| Letzte Änderung: | 02 Okt 2026 10:47 |
Nur für Mitarbeiter des Archivs: Kontrollseite des Eintrags