Create README.md

README.md

@@ -0,0 +1,102 @@
+---
+license: apache-2.0
+paper: https://arxiv.org/abs/2510.12670
+homepage: https://github.com/IBM/TerraCodec
+---
+
+# TerraCodec
+**Neural Compression for Earth Observation**
+
+TerraCodec (TEC) is a family of pretrained neural compression codecs for **multispectral Sentinel-2 satellite imagery**. The models compress optical Earth observation data using learned latent representations and entropy coding.
+
+Compared to classical codecs such as JPEG2000 or WebP, TerraCodec achieves **3–10× higher compression at comparable reconstruction quality** on multispectral satellite imagery. Temporal models further improve compression by exploiting redundancy across seasonal image sequences.
+
+📄 Paper: https://arxiv.org/abs/2510.12670  
+💻 GitHub: https://github.com/IBM/TerraCodec  
+
+---
+
+# Models
+
+| Model | Available Checkpoints | Description |
+|---|---|---|
+| `terracodec_v1_fp_s2l2a` | λ = 0.5, 2, 10, 40, 200 | Factorized-prior image codec. Smallest model and strong baseline for multispectral image compression. |
+| `terracodec_v1_elic_s2l2a` | λ = 0.5, 2, 10, 40, 200 | Enhanced entropy model with spatial and channel context, providing improved rate–distortion performance for image compression. |
+| `terracodec_v1_tt_s2l2a` | λ = 0.4, 1, 5, 20, 100, 200, 700 | Temporal Transformer codec modeling redundancy across seasonal multispectral image sequences. |
+| `flextec_v1_s2l2a` | **Single checkpoint** (quality = 1–16) | Flexible-rate temporal codec. One model supports multiple compression levels via token-based quality settings at inference time. |
+
+Lower λ/ quality → **higher compression**  
+Higher λ/ quality → **higher reconstruction quality**
+
+See the paper and GitHub for details.
+
+---
+
+# Installation
+
+```bash
+pip install terracodec
+```
+
+---
+
+# QuickStart
+
+```bash
+from terracodec import terracodec_v1_fp_s2l2a
+
+model = terracodec_v1_fp_s2l2a(
+    pretrained=True,
+    compression=10
+)
+
+# Fast Reconstruction
+reconstruction = model(inputs)
+
+# True Compression
+compressed = model.compress(inputs)
+reconstruction = model.decompress(**compressed)
+```
+
+---
+
+# Input Format
+
+
+| Codec type | Shape | Example |
+|---|---|---|
+| Image codecs | `[B, C, H, W]` | `[1, 12, 256, 256]` |
+| Temporal codecs | `[B, T, C, H, W]` | `[1, 4, 12, 256, 256]` |
+
+- **12 spectral bands** (Sentinel-2 L2A)
+- **Spatial size:** 256×256 recommended. TEC-FP accepts arbitrary sizes; all other models expect 256×256.
+- **Temporal models:** Models are pretrained on four seasonal frames but can process an arbitrary number of input timesteps at inference time. Using more frames increases the computational cost and therefore the required inference time.
+
+### Normalization
+
+Models were trained on [SSL4EO-S12 v1.1](https://huggingface.co/datasets/embed2scale/SSL4EO-S12-v1.1). Inputs should be standardized per spectral band using dataset statistics.
+For S2L2A:
+
+```python
+mean = torch.tensor([793.243, 924.863, 1184.553, 1340.936, 1671.402, 2240.082, 2468.412, 2563.243, 2627.704, 2711.071, 2416.714, 1849.625])
+std = torch.tensor([1160.144, 1201.092, 1219.943, 1397.225, 1400.035, 1373.136, 1429.170, 1485.025, 1447.836, 1652.703, 1471.002, 1365.307])
+```
+
+---
+
+## Citation
+
+```bibtex
+@article{terracodec2025,
+  title   = {TerraCodec: Neural Codecs for Earth Observation},
+  author  = {Costa Watanabe, Julen and Wittmann, Isabelle and Blumenstiel, Benedikt},
+  journal = {arXiv preprint arXiv:2510.12670},
+  year    = {2025}
+}
+```
+
+---
+
+## License
+
+Apache 2.0.