Temporal and Frequential Metric Learning for Time Series kNN Classification

Our approach for time series classification builds on two well-known methods in computer vision: local features are extracted from time series using a SIFT- based approach and a

In the experi- mental section, we evaluate the proposed kernel using SIFT features adapted to time series [2, 7] and show that it significantly outperforms the original algorithm

Our approach for time series classification builds on two well-known methods in computer vision: local features are extracted from time series using a SIFT- based approach and a

Second resubmission step: “PELICAN: deeP architecturE for the LIght Curve ANalysis” Johanna Pasquet, J´ erˆ ome Pasquet, Marc Chaumont and Dominique Fouchez. Astronomy

When dealing with high spatial resolution images, object-based approaches are generally used in order to exploit the spatial relationships of the data8. However, these

To compare the two methods relying on a two-steps ma- chine learning approach with our convolutional neural network and our Siamese network, we used the Base B1 with k-folds

The main idea is first to embed pairs of time series in a space whose dimensions are basic temporal metrics, and then to learn the metric combination function through a

With the arrival of SITS with high spatial resolution (HSR), it becomes necessary to use the spatial information held in these series, in order to either study the evolution of