Corpus ID: 53101390

ActionXPose: A Novel 2D Multi-view Pose-based Algorithm for Real-time Human Action Recognition

@article{Angelini2018ActionXPoseAN,
  title={ActionXPose: A Novel 2D Multi-view Pose-based Algorithm for Real-time Human Action Recognition},
  author={F. Angelini and Zeyu Fu and Yang Long and L. Shao and S. Naqvi},
  journal={ArXiv},
  year={2018},
  volume={abs/1810.12126}
}

F. Angelini, Zeyu Fu, +2 authors S. Naqvi

Published 2018

Computer Science, Engineering

ArXiv

We present ActionXPose, a novel 2D pose-based algorithm for posture-level Human Action Recognition (HAR. [...] Key Method The algorithm has real-time performance and it is robust to camera movings, subject proximity changes, viewpoint changes, subject appearance changes and provide high generalization degree. In fact, extensive simulations show that ActionXPose can be successfully trained using different datasets at once. State-of-the-art performance on popular datasets for posture-related HAR problems…Expand Abstract

View PDF on arXiv

Save to Library

Create Alert

Cite

Launch Research Feed