Efficient data-driven gap filling of satellite image time series using deep neural networks with partial convolutions. (arXiv:2208.08781v1 [cs.LG])

The abundance of gaps in satellite image time series often complicates the
application of deep learning models such as convolutional neural networks for
spatiotemporal modeling. Based on previous work in computer vision on image
inpainting, this paper shows how three-dimensional spatiotemporal partial
convolutions can be used as layers in neural networks to fill gaps in satellite
image time series. To evaluate the approach, we apply a U-Net-like model on
incomplete image time series of quasi-global carbon monoxide observations from
the …

arxiv data data-driven gap image lg networks neural networks satellite series time time series