Land cover prediction is essential for monitoring global environmental change. Unfortunately, traditional classification models are plagued by temporal variation and emergence of novel/unseen land cover classes in the prediction process. In this paper, we propose an LSTM-based spatiotemporal learning framework with a dual-memory structure. The dual-memory structure captures both long-term and short-term temporal variation patterns, and is updated incrementally to adapt the model to the ever-changing environment. Moreover, we integrate zero-shot learning to identify unseen classes even without labelled samples. Experiments on both synthetic and real-world datasets demonstrate the superiority of the proposed framework over multiple baselines in land cover prediction.
|Original language||English (US)|
|Title of host publication||KDD 2017 - Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining|
|Publisher||Association for Computing Machinery|
|Number of pages||10|
|State||Published - Aug 13 2017|
|Event||23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2017 - Halifax, Canada|
Duration: Aug 13 2017 → Aug 17 2017
|Name||Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining|
|Other||23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2017|
|Period||8/13/17 → 8/17/17|
Bibliographical noteFunding Information:
This work was funded by the NSF Awards 1029711, and Gordon and Betty Moore Foundation and the Belmont Forum/FACCE-JPI funded DEVIL project (NE/M021327/1). Access to  computing facilities was provided by NASA Earth Exchange and Minnesota Supercomputing Institute.
- Land cover
- Zero-short learning