TerraBytes II — ECCV 2026 Workshop TerraBytes — Towards global datasets and models for Earth observation ECCV 2026 Workshop Programme online 8 September 2026 · Malmö, Sweden Programme online · Second edition TerraBytes II The workshop on global datasets and models for Earth observation returns at ECCV 2026 — this September in Malmö, Sweden. 📅 8 September 2026 📍 Malmö, Sweden 📄 36 papers accepted · submissions closed Program Accepted papers Social Travel support About 2025 edition → About TerraBytes Earth observation (EO) presents unique challenges and opportunities that set it apart from other fields of machine learning and computer vision. EO data is abundant, repeatedly covering a large but bounded environment — our planet — and the co-location and evolution of these observations is a rich, multimodal, multitemporal source of information. Yet the distribution of EO data is non-stationary and spatially biased, with large parts of the world strongly under-represented. TerraBytes is an initiative to address these challenges at the intersection of data curation, data archiving, and representation learning, fostering a holistic discussion that covers every step from downlinked satellite data to training paradigms and downstream applications. After a successful first edition at ICML 2025, TerraBytes returns bigger at ECCV 2026. The programme and the 36 accepted papers are published below; remaining details will be confirmed on this page as they are settled. Program Full-day workshop, 8 September 2026. All times are local to Malmö (CEST). ⚠️ Preliminary programme This schedule is preliminary and may still change — including at short notice — due to ECCV 2026 conference organisation. Please check the exact timings on the day of the event via the official ECCV programme and on-site signage. Morning session 09:00–09:10 Opening remarks 09:10–09:50 Keynote 1 Talk title to be announced TBC Dr Noelia Jiménez Martínez · Earth Genome 09:50–10:10 Spotlights Best short papers 10:10–10:30 Coffee break 10:30–11:10 Keynote 2 “From Data Deluge to Digital Twins: Operating Europe’s Earth Observation Data Infrastructure” Lothar Wolf · EUMETSAT 11:10–12:00 Spotlights 1 Full papers 12:00–13:00 Lunch break Time TBC Lunch timing has not yet been confirmed by the ECCV organisation. Afternoon session 13:00–14:00 Spotlights 2 Full papers 14:00–15:00 Poster session All accepted papers 15:00–15:30 Coffee break 15:30–16:10 Keynote 3 “Contextualizing the pixels: more qualitative approaches to remote sensing” Dr Lina Eklund · Lund University 16:10–17:10 Panel session Towards Global Models Dr Kelsey Doerksen · Arizona State University and the African Climate & Development Initiative Georgia Channing · Hugging Face and the University of Oxford Prof Maria Antonia Brovelli · AI4Good, Politecnico di Milano 17:10–17:20 Closing remarks 18:30 Social Workshop social Time & venue TBC Off-site at a separate venue, following the workshop. Sign up on Luma for the address and updates. Evening · 8 September 2026 Workshop social After the closing remarks, join the TerraBytes community for an informal evening at a separate venue — a chance to carry the day’s conversations on over food and drinks. Starts 18:30. Exact time and venue are still being confirmed; the Luma page carries the address and any changes. Sign-up is required so we can plan numbers. Sign up on Luma 🎉 Travel Grant Award Travel support secured Asterisk Labs is sponsoring the TerraBytes II Travel Grant Award — helping authors who would otherwise be unable to travel present their work in Malmö. Applications are now closed, as author registration closes on Monday 10 August 2026. Everyone who registered their interest will be contacted by email. Awards prioritise early career scientists who are authors of accepted papers and would otherwise be unable to attend. Sponsored by Accepted papers 36 papers were accepted to TerraBytes II, grouped below by presentation format. Select a title to read its abstract. Accepted full-length papers may opt in to the archival proceedings; short papers are non-archival. Oral presentations23 Given spotlight talks during the morning and afternoon spotlight sessions, in addition to the poster session. 1 VLM2GeoVec: Toward Universal Multimodal Embeddings for Remote Sensing Emanuel Sanchez Aimar, Gulnaz Zhambulova, Fahad Shahbaz Khan, Yonghao Xu, Michael Felsberg Satellite imagery differs from natural images in viewpoint, resolution, scale variation, and the prevalence of small objects — demanding both region-level spatial reasoning and holistic scene understanding. Existing remote-sensing approaches are fragmented: dual-encoder retrieval models scale well but cannot interleave modalities, whereas generative assistants support grounding, yet are inefficient for retrieval. Benchmarks mirror this split: interleaved evaluations mainly target generative assistants, while cross-modal retrieval benchmarks target dual encoders. To bridge this gap, we introduce RSMEB, a unified remote sensing benchmark that evaluates cross-modal and interleaved retrieval across 21 tasks under a single ranking protocol, enabling comprehensive comparison of retrieval models on region- and geo-aware capabilities as well as conventional retrieval. As a strong reference baseline, we present VLM2GeoVec, an instruction-conditioned, single-encoder interleaving formulation tailored to remote sensing that packs image, text, bounding-box, and geo-coordinate tokens into one sequence and learns a unified embedding via contrastive training. Across RSMEB, VLM2GeoVec achieves 26.6% P@1 in region-caption retrieval (+25 percentage points), 32.5% in referring-expression retrieval (+19), and 17.8% in semantic geo-aware retrieval (>3× prior best), while remaining competitive in conventional scene classification and text–image retrieval in zero-shot settings. Together, the proposed suite and reference baseline standardize evaluation and deliver a unified embedder for scalable retrieval and region-/geo-aware grounding. Code, model checkpoints, and data will be released upon acceptance. PDF OpenReview forum → 2 SwissCrop25: A National Multi-Year Benchmark for Operational Crop Mapping Thomas Lauber, Mehmet Ozgur Turkoglu, Sélène Ledain, Helge Aasen Operational crop type mapping requires models that generalise across years, resolve fine-grained crop taxonomies, and distinguish cropland from surrounding landscapes. However, existing crop mapping datasets enable evaluation of these requirements only in isolation. We therefore introduce SwissCrop25, a national-scale crop mapping dataset spanning seven growing seasons (2019-2025), combining Sentinel-2 time series, daily temperature observations, a fine-grained 73 crop taxonomy with grassland management types, and 5 explicit non-crop land cover classes. To evaluate realistic deployment conditions, we define a five-fold leave-one-year-out protocol with joint cropland delineation and crop classification for benchmarking representative crop mapping architectures. Evaluating U-TAE, TSViT, and Galileo reveals differences between architectures hidden by conventional benchmarks. In this setting, domain-specific models outperform Galileo, a pretrained Earth observation foundation model, with TSViT achieving the best overall performance and a 12 pp mIoU advantage over U-TAE. TSViT's advantage increases with taxonomic specificity and class rarity, highlighting the importance of fine-grained evaluation. SwissCrop25 also exposes substantial interannual distribution shifts and shows that temperature priors improve robustness. Finally, in-season evaluation reveals a trade-off between models, with U-TAE providing stronger early-season performance on common crops and TSViT gaining an advantage later in the season through improved rare-class discrimination. SwissCrop25 provides a challenging testbed for evaluating crop mapping systems under realistic operational conditions and is publicly released. PDF OpenReview forum → 7 Spatial-Frequency Gated Swin Transformer for Cross-Sensor Remote Sensing Super-Resolution Md Aminur Hossain, Ayush V Patel, Yogesh Jethani, Sanjay K Singh, Biplab Banerjee Remote sensing single-image super-resolution aims to generate high-resolution imagery from low-resolution observations while preserving fine spatial structures such as roads, building boundaries, field edges, and land-cover transitions. Recent Swin Transformer-based models, including Swin2SR, provide strong spatial context modeling through shifted-window self-attention; however, their feed-forward networks remain generic channel-mixing modules and do not explicitly distinguish between low-frequency structural content and residual detail features. To address this limitation, we propose SFG-SwinSR, a Spatial-Frequency Gated Swin Transformer for cross-sensor remote sensing super-resolution. SFG-SwinSR replaces the standard feed-forward network in each Swin2SR transformer block with a lightweight Spatial-Frequency Gated Feed-Forward Network. The proposed module estimates low-frequency structural content through a depthwise low-pass branch, derives residual detail features by subtraction, refines them using a lightweight spatial branch, and adaptively reinjects useful details through a bottleneck gate. Experiments on three real cross-sensor benchmarks, SEN2VENµS, OLI2MSI, and SEN2NAIP, together with an auxiliary synthetic SpaceNet Challenge 3 setting, show consistent improvements across most evaluation settings and competitive performance against recent Swin-based baselines. The results demonstrate that spatial-frequency feature transformation inside transformer feed-forward networks provides an effective and lightweight inductive bias for structure-aware cross-sensor remote sensing super-resolution. PDF OpenReview forum → 9 Robust Satellite RPC Refinement via Bundle Adjustment with Season-Invariant Correspondences Roger Marí, Elías Masquil, Xavier Bou, Thibaud Ehret, Gabriele Facciolo Accurate refinement of Rational Polynomial Camera (RPC) models is essential for high-quality satellite image geolocation. In ground control point (GCP)-free multi-view pipelines, this refinement is commonly performed through bundle adjustment from automatically extracted image correspondences. However, conventional RPC bundle adjustment pipelines rely on handcrafted feature matching, which becomes unreliable in multi-date collections affected by seasonal, illumination, and land-cover changes. We develop an appearance-aware RPC refinement pipeline by systematically evaluating and integrating learned local matching and global image descriptors, together with a novel similarity-based image-pair selection strategy. This reduces redundant and error-prone matching while preserving the connectivity of the matching graph. Experiments on seasonally diverse WorldView-3 images show that the resulting pipeline improves GCP-free relative RPC refinement over open-source baselines, achieving lower geometric consistency errors while reducing matching time on collections with 39-42 views. These results demonstrate the effectiveness of the proposed system on WorldView-3 imagery and motivate broader evaluation across sensors and datasets. PDF OpenReview forum → 12 Locally Consistent Transductive Information Maximization for Few-Shot Remote Sensing Scene Classification Karim El Khoury, Benoît Gérin, Benoit Macq, Christophe De Vleeschouwer Remote sensing scene classification is increasingly relying on domain-specific foundation models. Moreover, transductive inference, which exploits the collective statistical structure of the entire unlabeled query set, appears to naturally match remote sensing pipelines where large images are routinely split into patches and inferred as a ba…