t-EVA: Time-Efficient t-SNE Video Annotation

Soroosh Poorgholi; Osman Semih Kayhan; Jan C. van Gemert

doi:10.1007/978-3-030-68799-1_12

t-EVA: Time-Efficient t-SNE Video Annotation

Soroosh Poorgholi^*, Osman Semih Kayhan, Jan C. van Gemert

^*Corresponding author for this work

Pattern Recognition and Bioinformatics

Research output: Chapter in Book/Conference proceedings/Edited volume › Conference contribution › Scientific › peer-review

1 Citation (Scopus)

Abstract

Video understanding has received more attention in the past few years due to the availability of several large-scale video datasets. However, annotating large-scale video datasets are cost-intensive. In this work, we propose a time-efficient video annotation method using spatio-temporal feature similarity and t-SNE dimensionality reduction to speed up the annotation process massively. Placing the same actions from different videos near each other in the two-dimensional space based on feature similarity helps the annotator to group-label video clips. We evaluate our method on two subsets of the ActivityNet (v1.3) and a subset of the Sports-1M dataset. We show that t-EVA (https://github.com/spoorgholi74/t-EVA ) can outperform other video annotation tools while maintaining test accuracy on video classification.

Original language	English
Title of host publication	Pattern Recognition. ICPR International Workshops and Challenges, 2021, Proceedings
Editors	Alberto Del Bimbo, Rita Cucchiara, Stan Sclaroff, Giovanni Maria Farinella, Tao Mei, Marco Bertini, Hugo Jair Escalante, Roberto Vezzani
Publisher	Springer
Pages	153-169
Number of pages	17
ISBN (Print)	9783030687984
DOIs	https://doi.org/10.1007/978-3-030-68799-1_12
Publication status	Published - 2021
Event	25th International Conference on Pattern Recognition Workshops, ICPR 2020 - Virtual, Online, Italy Duration: 10 Jan 2021 → 15 Jan 2021

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	12664 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	25th International Conference on Pattern Recognition Workshops, ICPR 2020
Country/Territory	Italy
City	Virtual, Online
Period	10/01/21 → 15/01/21

Keywords

Action recognition
t-SNE
Video annotation

Access to Document

10.1007/978-3-030-68799-1_12

Cite this

Poorgholi, S., Kayhan, O. S., & van Gemert, J. C. (2021). t-EVA: Time-Efficient t-SNE Video Annotation. In A. Del Bimbo, R. Cucchiara, S. Sclaroff, G. M. Farinella, T. Mei, M. Bertini, H. J. Escalante, & R. Vezzani (Eds.), Pattern Recognition. ICPR International Workshops and Challenges, 2021, Proceedings (pp. 153-169). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 12664 LNCS). Springer. https://doi.org/10.1007/978-3-030-68799-1_12

Poorgholi, Soroosh ; Kayhan, Osman Semih ; van Gemert, Jan C. / t-EVA : Time-Efficient t-SNE Video Annotation. Pattern Recognition. ICPR International Workshops and Challenges, 2021, Proceedings. editor / Alberto Del Bimbo ; Rita Cucchiara ; Stan Sclaroff ; Giovanni Maria Farinella ; Tao Mei ; Marco Bertini ; Hugo Jair Escalante ; Roberto Vezzani. Springer, 2021. pp. 153-169 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{9cdee259377f49c9be55f8f204be7a4c,

title = "t-EVA: Time-Efficient t-SNE Video Annotation",

abstract = "Video understanding has received more attention in the past few years due to the availability of several large-scale video datasets. However, annotating large-scale video datasets are cost-intensive. In this work, we propose a time-efficient video annotation method using spatio-temporal feature similarity and t-SNE dimensionality reduction to speed up the annotation process massively. Placing the same actions from different videos near each other in the two-dimensional space based on feature similarity helps the annotator to group-label video clips. We evaluate our method on two subsets of the ActivityNet (v1.3) and a subset of the Sports-1M dataset. We show that t-EVA (https://github.com/spoorgholi74/t-EVA ) can outperform other video annotation tools while maintaining test accuracy on video classification.",

keywords = "Action recognition, t-SNE, Video annotation",

author = "Soroosh Poorgholi and Kayhan, {Osman Semih} and {van Gemert}, {Jan C.}",

year = "2021",

doi = "10.1007/978-3-030-68799-1_12",

language = "English",

isbn = "9783030687984",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer",

pages = "153--169",

editor = "{Del Bimbo}, Alberto and Rita Cucchiara and Stan Sclaroff and Farinella, {Giovanni Maria} and Tao Mei and Marco Bertini and Escalante, {Hugo Jair} and Roberto Vezzani",

booktitle = "Pattern Recognition. ICPR International Workshops and Challenges, 2021, Proceedings",

note = "25th International Conference on Pattern Recognition Workshops, ICPR 2020 ; Conference date: 10-01-2021 Through 15-01-2021",

}

Poorgholi, S, Kayhan, OS & van Gemert, JC 2021, t-EVA: Time-Efficient t-SNE Video Annotation. in A Del Bimbo, R Cucchiara, S Sclaroff, GM Farinella, T Mei, M Bertini, HJ Escalante & R Vezzani (eds), Pattern Recognition. ICPR International Workshops and Challenges, 2021, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 12664 LNCS, Springer, pp. 153-169, 25th International Conference on Pattern Recognition Workshops, ICPR 2020, Virtual, Online, Italy, 10/01/21. https://doi.org/10.1007/978-3-030-68799-1_12

t-EVA: Time-Efficient t-SNE Video Annotation. / Poorgholi, Soroosh; Kayhan, Osman Semih ; van Gemert, Jan C.
Pattern Recognition. ICPR International Workshops and Challenges, 2021, Proceedings. ed. / Alberto Del Bimbo; Rita Cucchiara; Stan Sclaroff; Giovanni Maria Farinella; Tao Mei; Marco Bertini; Hugo Jair Escalante; Roberto Vezzani. Springer, 2021. p. 153-169 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 12664 LNCS).

Research output: Chapter in Book/Conference proceedings/Edited volume › Conference contribution › Scientific › peer-review

TY - GEN

T1 - t-EVA

T2 - 25th International Conference on Pattern Recognition Workshops, ICPR 2020

AU - Poorgholi, Soroosh

AU - Kayhan, Osman Semih

AU - van Gemert, Jan C.

PY - 2021

Y1 - 2021

N2 - Video understanding has received more attention in the past few years due to the availability of several large-scale video datasets. However, annotating large-scale video datasets are cost-intensive. In this work, we propose a time-efficient video annotation method using spatio-temporal feature similarity and t-SNE dimensionality reduction to speed up the annotation process massively. Placing the same actions from different videos near each other in the two-dimensional space based on feature similarity helps the annotator to group-label video clips. We evaluate our method on two subsets of the ActivityNet (v1.3) and a subset of the Sports-1M dataset. We show that t-EVA (https://github.com/spoorgholi74/t-EVA ) can outperform other video annotation tools while maintaining test accuracy on video classification.

AB - Video understanding has received more attention in the past few years due to the availability of several large-scale video datasets. However, annotating large-scale video datasets are cost-intensive. In this work, we propose a time-efficient video annotation method using spatio-temporal feature similarity and t-SNE dimensionality reduction to speed up the annotation process massively. Placing the same actions from different videos near each other in the two-dimensional space based on feature similarity helps the annotator to group-label video clips. We evaluate our method on two subsets of the ActivityNet (v1.3) and a subset of the Sports-1M dataset. We show that t-EVA (https://github.com/spoorgholi74/t-EVA ) can outperform other video annotation tools while maintaining test accuracy on video classification.

KW - Action recognition

KW - t-SNE

KW - Video annotation

UR - http://www.scopus.com/inward/record.url?scp=85103451699&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-68799-1_12

DO - 10.1007/978-3-030-68799-1_12

M3 - Conference contribution

AN - SCOPUS:85103451699

SN - 9783030687984

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 153

EP - 169

BT - Pattern Recognition. ICPR International Workshops and Challenges, 2021, Proceedings

A2 - Del Bimbo, Alberto

A2 - Cucchiara, Rita

A2 - Sclaroff, Stan

A2 - Farinella, Giovanni Maria

A2 - Mei, Tao

A2 - Bertini, Marco

A2 - Escalante, Hugo Jair

A2 - Vezzani, Roberto

PB - Springer

Y2 - 10 January 2021 through 15 January 2021

ER -

Poorgholi S, Kayhan OS , van Gemert JC. t-EVA: Time-Efficient t-SNE Video Annotation. In Del Bimbo A, Cucchiara R, Sclaroff S, Farinella GM, Mei T, Bertini M, Escalante HJ, Vezzani R, editors, Pattern Recognition. ICPR International Workshops and Challenges, 2021, Proceedings. Springer. 2021. p. 153-169. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-030-68799-1_12

t-EVA: Time-Efficient t-SNE Video Annotation

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this