TESSERA：用于地球表征与分析的地表光谱时序嵌入
TESSERA: Temporal Embeddings of Surface Spectra for Earth Representation and Analysis
TESSERA 用 Barlow Twins 加稀疏随机时序采样，让像素级嵌入对观测缺失保持不变，这种「对不规则采样鲁棒」的自监督目标可以直接迁移到视频或时序图像的自监督预训练中；它发布的 int8 全球嵌入和轻量适配头也为大规模检索提供了可复用的工程范式。
原文依据 TESSERA uses Barlow Twins and sparse random temporal sampling to enforce invariance to the selection of valid observations
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论光学和微波波段的卫星对地观测（EO）时间序列常因轨道模式和云层遮挡而不规则，虽然合成处理可以缓解这些问题，但会丢失关键的物候信息。为克服这一局限，我们提出 TESSERA，一个面向多模态（Sentinel-1/2）对地观测时间序列的像素级基础模型，能够学习鲁棒且标签高效的嵌入。训练时，TESSERA 采用 Barlow Twins 和稀疏随机时序采样，强制模型对有效观测的选择保持不变，并辅以两个关键正则化项：全局混洗以解耦空间邻域，以及基于混合的调节以实现极端稀疏下的不变性。我们发现，在多种分类、分割和回归任务上，TESSERA 嵌入以高标签效率达到最优精度，通常只需一个小型任务检测头和极少的计算。为促进普及、遵循 FAIR 原则并简化使用，我们发布了全球、年度、10 米分辨率的像素级 int8 嵌入，以及开放权重/代码和轻量适配头，为行星尺度的大规模检索和推理提供实用工具。所有代码和数据见 https://github.com/ucam-eo/tessera。
英文原文摘要 1263 字符
Satellite Earth-observation (EO) time series in the optical and microwave ranges are often irregular due to orbital patterns and cloud obstruction, and while compositing addresses these issues, it loses critical phenological information. To overcome this, we present TESSERA, a pixel-wise foundation model for multi-modal (Sentinel-1/2) EO time series that learns robust, label-efficient embeddings. During training, TESSERA uses Barlow Twins and sparse random temporal sampling to enforce invariance to the selection of valid observations, aided by two key regularizers: global shuffling to decorrelate spatial neighborhoods and mix-based regulation for invariance under extreme sparsity. We find that for diverse classification, segmentation, and regression tasks, TESSERA embeddings deliver state-of-the-art accuracy with high label efficiency, often requiring only a small task head and minimal computation. To democratize access, adhere to FAIR principles, and simplify use, we release global, annual, 10m, pixel-wise int8 embeddings together with open weights/code and lightweight adaptation heads, providing practical tooling for large-scale retrieval and inference at planetary scale. All code and data are available at https://github.com/ucam-eo/tessera.