A4 Conference proceedings

A Practical Hybrid Active Learning Approach for Human Pose Estimation


Open Access publication

Publication Details
Authors: Kaplan Sinan, Juvonen Joni, Lensu Lasse
Publisher: Springer Verlag (Germany): Series
Publication year: 2021
Language: English
Related journal or series: Lecture Notes in Computer Science
Title of parent publication: Structural, Syntactic, and Statistical Pattern Recognition
Journal acronym: LNCS
Volume number: 12644
ISBN: 978-3-030-73972-0
eISBN: 978-3-030-73973-7
ISSN: 0302-9743
eISSN: 1611-3349
JUFO level of this publication: 1
Open Access: Open Access publication
Location of the parallel saved publication: http://urn.fi/URN:NBN:fi-fe2021050629042

Abstract

Active learning (AL) has not received much attention in deep learning (DL) for human pose estimation. In this paper, a practical hybrid active learning strategy is proposed for training a human pose estimation model, and it is tested in an industrial online environment. The conducted experiments show that the active learning strategy to select diverse samples to be annotated outperforms the baseline method with random sampling. As a result, the strategy enables a significant improvement in the performance of pose estimation.


Last updated on 2021-06-05 at 13:26