Subspace Clustering for Action Recognition with Covariance Representations and Temporal Pruning

TL;DR

This paper introduces an unsupervised subspace clustering approach for human action recognition from skeletal data, utilizing covariance representations and temporal pruning to improve discriminability and performance.

Contribution

It presents a novel unsupervised subspace clustering method with covariance matrices and temporal pruning, outperforming existing unsupervised methods and rivaling supervised approaches.

Findings

Outperforms existing unsupervised methods

Achieves competitive results with supervised methods

Demonstrates effectiveness of covariance and temporal pruning

Abstract

This paper tackles the problem of human action recognition, defined as classifying which action is displayed in a trimmed sequence, from skeletal data. Albeit state-of-the-art approaches designed for this application are all supervised, in this paper we pursue a more challenging direction: Solving the problem with unsupervised learning. To this end, we propose a novel subspace clustering method, which exploits covariance matrix to enhance the action's discriminability and a timestamp pruning approach that allow us to better handle the temporal dimension of the data. Through a broad experimental validation, we show that our computational pipeline surpasses existing unsupervised approaches but also can result in favorable performances as compared to supervised methods.