Temporal Human Action Segmentation via Dynamic Clustering

TL;DR

This paper introduces a fast, unsupervised dynamic clustering algorithm for temporal human action segmentation, demonstrating state-of-the-art performance in both online and offline scenarios across various applications.

Contribution

The paper proposes a novel, generic dynamic clustering method that is unsupervised, fast, and applicable to multiple feature types for temporal action segmentation.

Findings

Achieves state-of-the-art results in online and offline settings.

Effective for various applications like robotics and patient monitoring.

Compatible with different feature types.

Abstract

We present an effective dynamic clustering algorithm for the task of temporal human action segmentation, which has comprehensive applications such as robotics, motion analysis, and patient monitoring. Our proposed algorithm is unsupervised, fast, generic to process various types of features, and applicable in both the online and offline settings. We perform extensive experiments of processing data streams, and show that our algorithm achieves the state-of-the-art results for both online and offline settings.