Unsupervised Discovery of Long-Term Spatiotemporal Periodic Workflows in Human Activities

TL;DR

This paper introduces a new benchmark dataset and a lightweight, training-free method for detecting and analyzing long-term periodic workflows in human activities, addressing a previously underexplored area.

Contribution

The paper presents the first benchmark for long-term periodic human activity workflows and a novel baseline method that outperforms existing approaches without training.

Findings

Benchmark challenges existing unsupervised and LLM-based methods.

Baseline method significantly outperforms competitors.

Deployment is effective without annotation or retraining.

Abstract

Periodic human activities with implicit workflows are common in manufacturing, sports, and daily life. While short-term periodic activities -- characterized by simple structures and high-contrast patterns -- have been widely studied, long-term periodic workflows with low-contrast patterns remain largely underexplored. To bridge this gap, we introduce the first benchmark comprising 580 multimodal human activity sequences featuring long-term periodic workflows. The benchmark supports three evaluation tasks aligned with real-world applications: unsupervised periodic workflow detection, task completion tracking, and procedural anomaly detection. We also propose a lightweight, training-free baseline for modeling diverse periodic workflow patterns. Experiments show that: (i) our benchmark presents significant challenges to both unsupervised periodic detection methods and zero-shot approaches based on powerful large language models (LLMs); (ii) our baseline outperforms competing methods by a substantial margin in all evaluation tasks; and (iii) in real-world applications, our baseline demonstrates deployment advantages on par with traditional supervised workflow detection approaches, eliminating the need for annotation and retraining. Our project page is https://sites.google.com/view/periodicworkflow.