# Unsupervised routine discovery in egocentric photo-streams

## TL;DR

This paper presents an unsupervised method to identify routine days from egocentric photo-streams, enabling better understanding of personal behavior patterns using wearable camera data.

## Contribution

It introduces a novel unsupervised outlier detection approach for recognizing routine-related days from first-person images, tested on real-world data from multiple users.

## Key findings

- Achieved 76% accuracy in identifying routine days.
- Attained 68% weighted F-Score across users.
- Demonstrated effectiveness of unsupervised models in behavior analysis.

## Abstract

The routine of a person is defined by the occurrence of activities throughout different days, and can directly affect the person's health. In this work, we address the recognition of routine related days. To do so, we rely on egocentric images, which are recorded by a wearable camera and allow to monitor the life of the user from a first-person view perspective. We propose an unsupervised model that identifies routine related days, following an outlier detection approach. We test the proposed framework over a total of 72 days in the form of photo-streams covering around 2 weeks of the life of 5 different camera wearers. Our model achieves an average of 76% Accuracy and 68% Weighted F-Score for all the users. Thus, we show that our framework is able to recognise routine related days and opens the door to the understanding of the behaviour of people.