Zero-Shot Motion Pattern Recognition from 4D Point-Clouds

Dariush Salami, Stephan Sigg

Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingsScientificpeer-review

84 Downloads (Pure)


We address a timely and relevant problem in signal processing: The recognition of patterns from spatial data in motion through a zero-shot learning scenario. We introduce a neural network architecture based on Siamese networks to recognize unseen classes of motion patterns. The approach uses a graph-based technique to achieve permutation invariance and also encodes moving point clouds into a representation space in a computationally efficient way. We evaluated the model on an open dataset with twenty-one gestures. The model out-performes state-of-the-art architectures with a considerable margin in four different settings in terms of accuracy while reducing the computational complexity up to 60 times.
Original languageEnglish
Title of host publication2021 IEEE 31st International Workshop on Machine Learning for Signal Processing, MLSP 2021
Number of pages6
ISBN (Electronic)978-1-7281-6338-3
ISBN (Print)978-1-6654-1184-4
Publication statusPublished - 15 Nov 2021
MoE publication typeA4 Conference publication
EventIEEE International Workshop on Machine Learning for Signal Processing - Gold Coast, Australia
Duration: 25 Oct 202128 Oct 2021
Conference number: 31


WorkshopIEEE International Workshop on Machine Learning for Signal Processing
Abbreviated titleMLSP
CityGold Coast
Internet address


  • Training
  • Computational modeling
  • Conferences
  • Neural networks
  • Computer architecture
  • Machine learning
  • Signal processing


Dive into the research topics of 'Zero-Shot Motion Pattern Recognition from 4D Point-Clouds'. Together they form a unique fingerprint.

Cite this