English

Combining Deep Learning Classifiers for 3D Action Recognition

Computer Vision and Pattern Recognition 2020-04-23 v1 Multimedia

Abstract

The popular task of 3D human action recognition is almost exclusively solved by training deep-learning classifiers. To achieve a high recognition accuracy, the input 3D actions are often pre-processed by various normalization or augmentation techniques. However, it is not computationally feasible to train a classifier for each possible variant of training data in order to select the best-performing subset of pre-processing techniques for a given dataset. In this paper, we propose to train an independent classifier for each available pre-processing technique and fuse the classification results based on a strict majority vote rule. Together with a proposed evaluation procedure, we can very efficiently determine the best combination of normalization and augmentation techniques for a specific dataset. For the best-performing combination, we can retrospectively apply the normalized/augmented variants of input data to train only a single classifier. This also allows us to decide whether it is better to train a single model, or rather a set of independent classifiers.

Keywords

Cite

@article{arxiv.2004.10314,
  title  = {Combining Deep Learning Classifiers for 3D Action Recognition},
  author = {Jan Sedmidubsky and Pavel Zezula},
  journal= {arXiv preprint arXiv:2004.10314},
  year   = {2020}
}

Comments

Submitted to Pattern Recognition Letters