Spatio-Temporal Alignments: Optimal transport through space and time

Abstract : Comparing data defined over space and time is notoriously hard, because it involves quantifying both spatial and temporal variability, while at the same time taking into account the chronological structure of data. Dynamic Time Warping (DTW) computes an optimal alignment between time series in agreement with the chronological order, but is inherently blind to spatial shifts. In this paper, we propose Spatio-Temporal Alignments (STA), a new differentiable formulation of DTW, in which spatial differences between time samples are accounted for using regularized optimal transport (OT). Our temporal alignments are handled through a smooth variant of DTW called soft-DTW, for which we prove a new property: soft-DTW increases quadrati-cally with time shifts. The cost matrix within soft-DTW that we use are computed using unbalanced OT, to handle the case in which observations are not normalized probabilities. Experiments on handwritten letters and brain imaging data confirm our theoretical findings and illustrate the effectiveness of STA as a dissimilarity for spatio-temporal data.
Document type :
Preprints, Working Papers, ...
Complete list of metadatas

Cited literature [14 references]  Display  Hide  Download

https://hal.inria.fr/hal-02309340
Contributor : Hicham Janati <>
Submitted on : Wednesday, October 9, 2019 - 11:08:04 AM
Last modification on : Friday, October 11, 2019 - 1:17:06 AM

File

arxiv___aistats20.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-02309340, version 1

Citation

Hicham Janati, Marco Cuturi, Alexandre Gramfort. Spatio-Temporal Alignments: Optimal transport through space and time. 2019. ⟨hal-02309340⟩

Share

Metrics

Record views

24

Files downloads

88