Comparison of manual, machine learning, and hybrid methods for video annotation to extract parental care data - Journal article summary

Aug 4, 2026

Efficient measurement of parental care in wild animal ecology is central to understanding animal ecology and evolution, yet it is often labour- and time-intensive. This short communication evaluates how manual expert annotation, crowd-sourcing, automatic detection using DeepMeerkat, and a hybrid approach extract parental provisioning from field video recordings of wild house sparrows. Automatic data correlated with experts and predicted brood survival, but it showed biased estimates from nonvisitation detections. Crowd-sourcing and hybrid methods matched expert accuracy, with hybrid requiring ~20% of manual annotation time.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages12

Reading Time30 min

AccessFree to download

Introduction

Methods comparison for extracting parental provisioning from videos

Results: correlation, bias, and biological meaning

Implications for cost-effective annotation workflows

This is a repository copy of Comparison of manual, machine learning, and hybrid methods for video annotation to extract parental care data. White Rose Research Online URL for this paper: [https://eprints.whiterose.ac.uk/206736/](https://eprints.whiterose.ac.uk/206736/) Version: Published Version Article: Chan, [A.H.H. orcid.org/0000-0002-5405-7155](A.H.H. orcid.org/0000-0002-5405-7155) , Liu, J., Burke, [T. orcid.org/0000-0003-3848-](T. orcid.org/0000-0003-3848-)[ ](T. orcid.org/0000-0003-3848-)[1244 et al.](1244 et al.) (2 more authors) (2023) Comparison of manual, machine learning, and hybrid methods for video annotation to extract parental care data. Journal of Avian Biology. e03167 . ISSN 0908-8857 [https://doi.org/10.1111/jav.03167](https://doi.org/10.1111/jav.03167) Reuse This article is distributed under the terms of the Creative Commons Attribution (CC BY) licence. This licence allows you to distribute, remix, tweak, and build upon the work, even commercially, as long as you credit the authors for the original work. More information and the full terms of the licence here: [https://creativecommons.org/licenses/](https://creativecommons.org/licenses/) Takedown If you consider content in White Rose Research Online to be in breach of UK law, please notify us by emailing [eprints@whiterose.ac.uk](eprints@whiterose.ac.uk) including the URL of the record and the reason for the withdrawal request. [eprints@whiterose.ac.uk](eprints@whiterose.ac.uk)[ ](eprints@whiterose.ac.uk)[https://eprints.whiterose.ac.uk/](https://eprints.whiterose.ac.uk/) JOURNAL OF AVIAN BIOLOGY Short communication Comparison of manual, machine learning, and hybrid methods for video annotation to extract parental care data Alex Hoi Hang Chan✉1,2,3, Jingqi Liu1, Terry Burke4, William D. Pearse1,* and Julia Schroeder1,* 1Department of Life Sciences, Imperial College London, Ascot, UK 2Centre for the Advanced Study of Collective Behaviour, University of Konstanz, Konstanz, Germany 3Max Planck Institute of Animal Behaviour, Radolfzell, Germany 4Department of Animal and Plant Sciences, University of Sheffield, Sheffield, UK Correspondence: Alex Hoi Hang Chan ([thealex.chan2@gmail.com](thealex.chan2@gmail.com)) Journal of Avian Biology 2023: e03167 doi: 10.1111/jav.03167 Subject Editor: Wesley Hochachka Editor-in-Chief: Staffan Bensch Accepted 16 October 2023 [www.avianbiology.org](www.avianbiology.org) Measuring parental care behaviour in the wild is central to the study of animal ecology and evolution, but it is often labour- and time-intensive. Efficient open-source tools have recently emerged that allow animal behaviour to be quantified from videos using machine learning and computer vision techniques, but there is limited appraisal of how these tools perform compared to traditional methods. To gain insight into how different methods perform in extracting data from videos taken in the field, we compared estimates of the parental provisioning rate of wild house sparrows Passer domesticus from video recordings. We compared four methods: manual annotation by experts, crowd-sourcing, automatic detection based on the open-source software DeepMeerkat, and a hybrid annotation method. We found that the data collected by the automatic method correlated with expert annotation (r = 0.62) and further show that these data are biologically meaningful as they predict brood survival. However, the automatic method produced largely biased estimates due to the detection of nonvisitation events, while the crowd-sourcing and hybrid annotation produced estimates that are equivalent to expert annotation. The hybrid annotation method takes approximately 20% of annotation time compared to manual annotation, making it a more cost-effective way to collect data from videos. We provide a successful case study of how different approaches can be adopted and evaluated with a pre-existing dataset, to make informed decisions on the best w

FAQs

Which four video annotation methods were compared in the study?

The study compared manual expert annotation, crowd-sourcing, automatic detection using DeepMeerkat, and a hybrid annotation method.

How did the automatic method perform relative to expert annotations?

Automatic detections correlated with expert annotation (r = 0.62) and predicted brood survival, but produced largely biased estimates due to detection of nonvisitation events.

Why is the hybrid annotation approach considered more efficient?

The hybrid method required about 20% of the annotation time compared with manual annotation while producing estimates equivalent to expert annotation.

Aug 4, 2026

Upload by Lucas Martin