Filling observational gaps

Last updated Jul 7, 2022 Edit Source

See AI/Computer Vision/Inpainting and restoration

• #PAPER Inpainting of Remote Sensing SST Images With Deep Convolutional Generative Adversarial Network (Dong 2018)
• #PAPER Unsupervised Inpainting for Occluded Sea Surface Temperature Sequences (Yin et al., 2019)
  • #CODE https://github.com/yuan-yin/UNISST
  • Unsupervised Spatiotemporal Data Inpainting
• #PAPER Artificial intelligence reconstructs missing climate information (Kadow 2020)
  • https://ecmwfevents.com/assets/posters/machine-learning-ws-kadow1601285092.pdf
  • #CODE https://github.com/FREVA-CLINT/climatereconstructionAI
• #PAPER Predicting into unknown space? Estimating the area of applicability of spatial prediction models (Meyer 2020)
  • #CODE https://github.com/HannaMeyer/AOA_CaseStudy
• #PAPER CLIMFILL v0.9: a framework for intelligently gap filling Earth observations (Bessenbacher 2022)
  • #CODE https://github.com/climachine/climfill
  • CLIMFILL fills gaps in gridded geoscientific observational data by taking into account spatial neighborhood, temporal context and multivariate dependencies. It takes a multivariate dataset with any number and pattern of missing values per variable and returns the dataset with all missing points replaced by estimates
• #PAPER Positional Encoder Graph Neural Networks for Geographic Data (Klemmer 2022)
  • #CODE https://github.com/konstantinklemmer/pe-gnn