English

SL-DML: Signal Level Deep Metric Learning for Multimodal One-Shot Action Recognition

Computer Vision and Pattern Recognition 2020-10-20 v4

Abstract

Recognizing an activity with a single reference sample using metric learning approaches is a promising research field. The majority of few-shot methods focus on object recognition or face-identification. We propose a metric learning approach to reduce the action recognition problem to a nearest neighbor search in embedding space. We encode signals into images and extract features using a deep residual CNN. Using triplet loss, we learn a feature embedding. The resulting encoder transforms features into an embedding space in which closer distances encode similar actions while higher distances encode different actions. Our approach is based on a signal level formulation and remains flexible across a variety of modalities. It further outperforms the baseline on the large scale NTU RGB+D 120 dataset for the One-Shot action recognition protocol by 5.6%. With just 60% of the training data, our approach still outperforms the baseline approach by 3.7%. With 40% of the training data, our approach performs comparably well to the second follow up. Further, we show that our approach generalizes well in experiments on the UTD-MHAD dataset for inertial, skeleton and fused data and the Simitate dataset for motion capturing data. Furthermore, our inter-joint and inter-sensor experiments suggest good capabilities on previously unseen setups.

Keywords

Cite

@article{arxiv.2004.11085,
  title  = {SL-DML: Signal Level Deep Metric Learning for Multimodal One-Shot Action Recognition},
  author = {Raphael Memmesheimer and Nick Theisen and Dietrich Paulus},
  journal= {arXiv preprint arXiv:2004.11085},
  year   = {2020}
}

Comments

8 pages, 6 figures, 7 tables

R2 v1 2026-06-23T15:02:57.926Z