Research map / Time series / Foundation models for time series
Satellite image time series
Classify and analyze satellite image sequences with datasets and temporal encoders.
28 accepted papers at ICML, NeurIPS, ICLR, CVPR and AAAI (2019–2026) · an approach in Foundation models for time series, keyword cluster “contrastive · classification · supervised”
State History Tried Preprints Who Export
Where it stands
Not enough benchmark data Fewer than four papers report the same benchmark, so there is no record to follow.
15 papers in 2023–24 and 7 in 2025–26 (at ICML, NeurIPS, ICLR, CVPR and AAAI); its share of the topic shrank, from 14.0% to 2.8%.
How this is decided
A benchmark is tracked when at least four papers of this approach report their own method on the same dataset and metric in at least two different years. Its record is the best number reported up to 2024 compared with the best reported in 2025–26: still improving means at least 1% better, flat means at most 0.2% better with at least two papers from 2025–26 trying. Numbers on a different scale from the rest (a fraction among percentages) and far outliers are left out. Results come from different papers with different setups, so they are signals, not head-to-head comparisons. The share of papers is counted at ICML, NeurIPS, ICLR, CVPR and AAAI.
How it got here
Most cited paper of each year
- 2019 · DeepDPM: Dynamic Population Mapping via Deep Neural Network (AAAI 2019 · 28 citations)
- 2020 · Satellite Image Time Series Classification With Pixel-Set Encoders and Temporal Self-Attention (CVPR 2020 · 144 citations)
- 2021 · The Multi-Temporal Urban Development SpaceNet Dataset (CVPR 2021 · 93 citations)
- 2022 · DynamicEarthNet: Daily Multi-Spectral Satellite Dataset for Semantic Change Segmentation (CVPR 2022 · 118 citations)
- 2023 · ViTs for SITS: Vision Transformers for Satellite Image Time Series (CVPR 2023 · 102 citations)
- 2024 · Multi-modal Learning for Geospatial Vegetation Forecasting (CVPR 2024 · 22 citations)
- 2025 · Exact: Exploring Space-Time Perceptive Clues for Weakly Supervised Satellite Image Time Series Semantic Segmentation (CVPR 2025 · 3 citations)
- 2026 · TESSERA: Temporal Embeddings of Surface Spectra for Earth Representation and Analysis (CVPR 2026 · 41 citations)
Earliest papers here
- DeepDPM: Dynamic Population Mapping via Deep Neural Network (AAAI 2019 · 28 citations)
- Satellite Image Time Series Classification With Pixel-Set Encoders and Temporal Self-Attention (CVPR 2020 · 144 citations)
- The Multi-Temporal Urban Development SpaceNet Dataset (CVPR 2021 · 93 citations)
What has been tried
24 papers here report 217 measured changes (a component added, removed or swapped, compared with a baseline): helped 77%, no clear effect 6%, hurt 14%, mixed 3%. Papers mostly publish changes that helped, so "helped" is over-represented.
Ablations in recent papers
| What was changed | Measured on | Result | Effect | Paper |
|---|---|---|---|---|
| use meteorological weather conditioning | GreenEarthNet OOD-t · R2 | Helped | Multi-modal Learning for Geospatial Vegetation Forecasting (CVPR 2024 · 22 citations) | |
| Add the hierarchical encoder | 44 downstream tasks · task-specific evaluation metrics | Helped | PowerPM: Foundation Model for Power Systems (NeurIPS 2024 · 9 citations) | |
| add the gaze MFCC module to the spacing-only model | Real-HFD · F1 | 93.9 -> 94.5 | Helped | Hidden Follower Detection: How Is the Gaze-Spacing Pattern Embodied in Frequency Domain? (AAAI 2024 · 5 citations) |
| add patch alignment module | CYP, corn · RMSE | 17.198 -> 16.192 | Helped | Efficient Representation Learning of Satellite Image Time Series and Their Fusion for Spatiotemporal Applications (AAAI 2024 · 4 citations) |
| Add CondTSF to MTT | ExchangeRate · MAE | 0.778±0.084 -> 0.195±0.007 | Helped | CondTSF: One-line Plugin of Dataset Condensation for Time Series Forecasting (NeurIPS 2024 · 4 citations) |
| Add clue-based contrastive learning Lcbl | PASTIS train · mIoU | 69.5 -> 72.3 | Helped | Exact: Exploring Space-Time Perceptive Clues for Weakly Supervised Satellite Image Time Series Semantic Segmentation (CVPR 2025 · 3 citations) |
| Add Top-K estimation to MME | GeoPlant PA test · F1s | 33.8 -> 35.9 | Helped | GeoPlant: Spatial Plant Species Prediction Dataset (NeurIPS 2024 · 3 citations) |
| add data augmentation in the sequential modality | Longliner · F1 | 0.53 -> 0.97 | Helped | GIST: Gear Type Identification by Spatiotemporal Trajectory Transformation for Monitoring Fisheries (AAAI 2025 · 1 citation) |
| add contrastive loss | METRLA, PEMSBAY, PEMSD7(M), PEMS04, PEMS07, PEMS08 · RMSE | 6.16 -> 6.02; 3.65 -> 3.51; 5.18 -> 5.11; 29.81 -> 29.64; 33.16 -> 32.51; 23.31 -> 23.10 | Helped | How Different from the Past? Spatio-Temporal Time Series Forecasting with Self-Supervised Deviation Learning (NeurIPS 2025 · 0 citations) |
| Represent 16-bit pixels as two 8-bit channels (MSB and LSB) instead of direct 16-bit input | AstroCompress datasets · compression ratio | Mixed | AstroCompress: A benchmark dataset for multi-purpose compression of astronomical data (ICLR 2025 · 0 citations) |
Recent preprints
Loading recent arXiv preprints…
Who publishes on it
Institutions on the most papers of this approach since 2023 (of 22 papers; from OpenAlex and the papers' affiliations).
- Hong Kong University of Science and Technology · 4 papers
- University of Pittsburgh · 2 papers
- University of Cambridge · 2 papers
- Inria · 2 papers
Use it
Download BibTeX of 28 key papersOpen the topic in the interactive map
Get alerts. Describe your idea in scime atlas and save it (free scime account): you get a weekly email when new preprints land close to it.
Related approaches
Same keyword cluster
- Time series representation learning (115 papers)
- Time series classification (61 papers)
Elsewhere in the topic
- LLM-based time series forecasting (90 papers)
- Time series foundation models (71 papers)
- Time-series language reasoning (44 papers)
Benchmark numbers, ablations and baseline numbers are extracted by AI from each paper's full text and can be wrong; the paper is the authority. Data: results index of 2026-11-02, citation counts from OpenAlex (2026-10-09).