Unsupervised Routine Discovery in Egocentric Photo-Streams

Estefanía Talavera; Nicolai Petkov; Petia Radeva

doi:10.1007/978-3-030-29888-3_47

Unsupervised Routine Discovery in Egocentric Photo-Streams

Estefanía Talavera^*, Nicolai Petkov, Petia Radeva

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

1 Downloads (Pure)

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.

Original language	English
Title of host publication	Computer Analysis of Images and Patterns
Subtitle of host publication	18th International Conference, CAIP 2019, Salerno, Italy, September 3–5, 2019, Proceedings
Editors	Mario Vento, Gennaro Percannella
Place of Publication	Cham
Publisher	Springer
Pages	576-588
Volume	Part I
ISBN (Electronic)	978-3-030-29888-3
ISBN (Print)	978-3-030-29887-6
DOIs	https://doi.org/10.1007/978-3-030-29888-3_47
Publication status	Published - 2019
Externally published	Yes
Event	18th International Conference on Computer Analysis of Images and Patterns, CAIP 2019 - Salerno, Italy Duration: 3 Sep 2019 → 5 Sep 2019 Conference number: 18

Publication series

Name	Lecture Notes in Computer Science
Publisher	Springer
Volume	11678
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	18th International Conference on Computer Analysis of Images and Patterns, CAIP 2019
Abbreviated title	CAIP 2019
Country/Territory	Italy
City	Salerno
Period	3/09/19 → 5/09/19

Keywords

Routine discovery
Lifestyle
Egocentric vision
Behaviour analysis

Access to Document

10.1007/978-3-030-29888-3_47

Talavera2019unsupervisedAccepted author version, 1.28 MB

Cite this

Talavera, E., Petkov, N., & Radeva, P. (2019). Unsupervised Routine Discovery in Egocentric Photo-Streams. In M. Vento, & G. Percannella (Eds.), Computer Analysis of Images and Patterns: 18th International Conference, CAIP 2019, Salerno, Italy, September 3–5, 2019, Proceedings (Vol. Part I, pp. 576-588). (Lecture Notes in Computer Science; Vol. 11678). Springer. https://doi.org/10.1007/978-3-030-29888-3_47

Talavera, Estefanía ; Petkov, Nicolai ; Radeva, Petia. / Unsupervised Routine Discovery in Egocentric Photo-Streams. Computer Analysis of Images and Patterns: 18th International Conference, CAIP 2019, Salerno, Italy, September 3–5, 2019, Proceedings. editor / Mario Vento ; Gennaro Percannella. Vol. Part I Cham : Springer, 2019. pp. 576-588 (Lecture Notes in Computer Science).

@inproceedings{fa7b30e3935241b7bf0cb5ce8b34b089,

title = "Unsupervised Routine Discovery in Egocentric Photo-Streams",

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.",

keywords = "Routine discovery, Lifestyle, Egocentric vision, Behaviour analysis",

author = "Estefan{\'i}a Talavera and Nicolai Petkov and Petia Radeva",

year = "2019",

doi = "10.1007/978-3-030-29888-3_47",

language = "English",

isbn = "978-3-030-29887-6",

volume = "Part I",

series = "Lecture Notes in Computer Science",

publisher = "Springer",

pages = "576--588",

editor = "Mario Vento and Gennaro Percannella",

booktitle = "Computer Analysis of Images and Patterns",

note = "18th International Conference on Computer Analysis of Images and Patterns, CAIP 2019, CAIP 2019 ; Conference date: 03-09-2019 Through 05-09-2019",

}

Talavera, E, Petkov, N & Radeva, P 2019, Unsupervised Routine Discovery in Egocentric Photo-Streams. in M Vento & G Percannella (eds), Computer Analysis of Images and Patterns: 18th International Conference, CAIP 2019, Salerno, Italy, September 3–5, 2019, Proceedings. vol. Part I, Lecture Notes in Computer Science, vol. 11678, Springer, Cham, pp. 576-588, 18th International Conference on Computer Analysis of Images and Patterns, CAIP 2019, Salerno, Italy, 3/09/19. https://doi.org/10.1007/978-3-030-29888-3_47

Unsupervised Routine Discovery in Egocentric Photo-Streams. / Talavera, Estefanía; Petkov, Nicolai; Radeva, Petia.

Computer Analysis of Images and Patterns: 18th International Conference, CAIP 2019, Salerno, Italy, September 3–5, 2019, Proceedings. ed. / Mario Vento; Gennaro Percannella. Vol. Part I Cham : Springer, 2019. p. 576-588 (Lecture Notes in Computer Science; Vol. 11678).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - Unsupervised Routine Discovery in Egocentric Photo-Streams

AU - Talavera, Estefanía

AU - Petkov, Nicolai

AU - Radeva, Petia

N1 - Conference code: 18

PY - 2019

Y1 - 2019

N2 - 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.

AB - 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.

KW - Routine discovery

KW - Lifestyle

KW - Egocentric vision

KW - Behaviour analysis

U2 - 10.1007/978-3-030-29888-3_47

DO - 10.1007/978-3-030-29888-3_47

M3 - Conference contribution

SN - 978-3-030-29887-6

VL - Part I

T3 - Lecture Notes in Computer Science

SP - 576

EP - 588

BT - Computer Analysis of Images and Patterns

A2 - Vento, Mario

A2 - Percannella, Gennaro

PB - Springer

CY - Cham

T2 - 18th International Conference on Computer Analysis of Images and Patterns, CAIP 2019

Y2 - 3 September 2019 through 5 September 2019

ER -

Talavera E, Petkov N, Radeva P. Unsupervised Routine Discovery in Egocentric Photo-Streams. In Vento M, Percannella G, editors, Computer Analysis of Images and Patterns: 18th International Conference, CAIP 2019, Salerno, Italy, September 3–5, 2019, Proceedings. Vol. Part I. Cham: Springer. 2019. p. 576-588. (Lecture Notes in Computer Science). doi: 10.1007/978-3-030-29888-3_47

Unsupervised Routine Discovery in Egocentric Photo-Streams

Abstract

Publication series

Conference

Keywords

Access to Document

Fingerprint

Cite this