TI-PREGO: Chain of Thought and In-Context Learning for Online Mistake Detection in PRocedural EGOcentric Videos

TL;DR

This paper introduces TI-PREGO, a dual-branch system combining step recognition and future step anticipation using LLMs to detect procedural mistakes online in egocentric videos, addressing open-set errors without prior failure examples.

Contribution

The paper proposes a novel dual-branch architecture utilizing LLMs for online mistake detection in egocentric videos, advancing open-set error recognition in procedural tasks.

Findings

Effective detection of procedural mistakes in real-time

Robustness demonstrated across multiple datasets

Outperforms existing state-of-the-art models

Abstract

Identifying procedural errors online from egocentric videos is a critical yet challenging task across various domains, including manufacturing, healthcare, and skill-based training. The nature of such mistakes is inherently open-set, as unforeseen or novel errors may occur, necessitating robust detection systems that do not rely on prior examples of failure. Currently, however, no technique effectively detects open-set procedural mistakes online. We propose a dual branch architecture to address this problem in an online fashion: one branch continuously performs step recognition from the input egocentric video, while the other anticipates future steps based on the recognition module's output. Mistakes are detected as mismatches between the currently recognized action and the action predicted by the anticipation module. The recognition branch takes input frames, predicts the current action, and aggregates frame-level results into action tokens. The anticipation branch, specifically, leverages the solid pattern-matching capabilities of Large Language Models (LLMs) to predict action tokens based on previously predicted ones. Given the online nature of the task, we also thoroughly benchmark the difficulties associated with per-frame evaluations, particularly the need for accurate and timely predictions in dynamic online scenarios. Extensive experiments on two procedural datasets demonstrate the challenges and opportunities of leveraging a dual-branch architecture for mistake detection, showcasing the effectiveness of our proposed approach. In a thorough evaluation including recognition and anticipation variants and state-of-the-art models, our method reveals its robustness and effectiveness in online applications.