Deep Metric Learners for Human Activity Recognition

Here we performe a comparative evaluation of Siamese Neural Networks, Triplet Networks and Matching Networks for the task of Human Activity Recognition using two HAR datasets: SelfBACK and PAMAP2

Publication: "Human activity recognition with deep metric learners" at the 27th International conference on case-based reasoning workshop (ICCBR-WS19), co-located with the 27th International conference on case-based reasoning (ICCBR19)

Citation:

@inproceedings{martin2020human,
title={Human activity recognition with deep metric learners.},
author={Martin, Kyle and Wijekoon, Anjana and Wiratunga, Nirmalie},
year={2020},
organization={CEUR Workshop Proceedings}
}'''