• English
    • Welsh
  • English 
    • English
    • Welsh
  • Login
Search DSpace:
  • Home
  • Research at Cardiff Met
  • Library Services
  • Contact Us
View item 
  • DSpace home
  • Cardiff School of Technologies
  • School of Technologies Research
  • View item
  • DSpace home
  • Cardiff School of Technologies
  • School of Technologies Research
  • View item
JavaScript is disabled for your browser. Some features of this site may not work without it.

Online tracking of ants based on deep association metrics: method, dataset and evaluation

Thumbnail
View/open
Author's post-print (5.184Mb)
Author
Cao, Xiaoyan
Guo, Shihui
Lin, Juncong
Zhang, Wenshu
Liao, Minghong
Date
2020-02-21
Acceptance date
2020-01-23
Type
Article
Publisher
Elsevier
ISSN
0031-3203
Embargoed until
2021-02-21
Metadata
Show full item record
Abstract
Tracking movement of insects in a social group (such as ants) is challenging, because the individuals are not only similar in appearance but also likely to perform intensive body contact and sudden movement adjustment (start/stop, direction changes). To address this challenge, we introduce an online multi-object tracking framework that combines both the motion and appearance information of ants. We obtain the appearance descriptors by using the ResNet model for offline training on a small (N=50) sample dataset. For online association, a cosine similarity metric computes the matching degree between historical appearance sequences of the trajectory and the current detection. We validate our method in both indoor (lab setup) and outdoor video sequences. The results show that our model obtains 99.3% ± 0.5% MOTA and 91.9% ± 2.1% MOTP across 24,050 testing samples in five indoor sequences, with real-time tracking performance. In an outdoor sequence, we achieve 99.3% MOTA and 92.9% MOTP across 22,041 testing samples. The datasets and code are made publicly available for future research in relevant domains.
Journal/conference proceeding
Pattern Recognition;
Citation
Cao, X., Guo, S., Lin, J., Zhang, W. and Liao, M. (2020) 'Online tracking of ants based on deep association metrics: method, dataset and evaluation', Pattern Recognition, p.107233. https://doi.org/10.1016/j.patcog.2020.107233
URI
http://hdl.handle.net/10369/11173
DOI
https://doi.org/10.1016/j.patcog.2020.107233
Description
Article published in Pattern Recognition available at https://doi.org/10.1016/j.patcog.2020.107233
Rights
http://www.rioxx.net/licenses/under-embargo-all-rights-reserved
Sponsorship
Cardiff Metropolitan University (Grant ID: Cardiff Metropolian (Internal))
Collections
  • School of Technologies Research [147]

Related items

Showing items related by title, author, subject and abstract.

  • Thumbnail

    AN IDENTIFICATION AND CRITICAL ANALYSIS OF FACTORS INFLUENCING AN ATHLETE’S PROGRESSION TO ELITE STATUS: A BRITISH PERSPECTIVE 

    Haslett, Michael (University of Wales Institute Cardiff, 2012)
    The present study sought to investigate the many contextual factors that play a role in athlete development and the achievement of expertise in Track and Field, specifically focusing on the perspective from Great Britain. ...
  • Thumbnail

    Evaluating and developing the key determinants of push-start performance in bobsleigh 

    Condliffe, Robert (Cardiff Metropolitan University, 2018)
    It is a common belief in bobsleigh that the push-start is a vital aspect of successful performance. Therefore, British Bobsleigh places a heavy emphasis on the use of field-based performance testing to assist with ...
  • Thumbnail

    Effects of situational variables on the physical activity profiles of elite soccer players in different score line states 

    Redwood-Brown, Athalie; O'Donoghue, Peter; Nevill, Alan; Saward, Chris; Dyer, Nicholas; Sunderland, Caroline (Wiley, 2018-07-28)
    The aim of this study were to investigate the effects of playing position, pitch location, team ability and opposition ability on the physical activity profiles of English premier league soccer players in difference score ...

Browse

DSpace at Cardiff MetCommunities & CollectionsBy issue dateAuthorsTitlesSubjectsThis collectionBy issue dateAuthorsTitlesSubjects

My Account

Login

Statistics

Most Popular ItemsStatistics by CountryMost Popular Authors

DSpace software copyright © 2002-2015  DuraSpace
Contact us | Send feedback | Administrator