Chimp&See: an online citizen science platform for large-scale, remote video camera trap annotation of chimpanzee behaviour, demography and individual identification

Department of Primatology, Max Planck Institute for Evolutionary Anthropology, Leipzig, Germany
DOI
10.7287/peerj.preprints.1792v1
Subject Areas
Animal Behavior, Conservation Biology, Ecology, Evolutionary Studies, Zoology
Keywords
video, Citizen science, camera trap, internet, chimpanzee
Copyright
© 2016 Arandjelovic et al.
Licence
This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. For attribution, the original author(s), title, publication source (PeerJ Preprints) and either DOI or URL of the article must be cited.
Cite this article
Arandjelovic M, Stephens CR, McCarthy MS, Dieguez P, Kalan AK, Maldonado N, Boesch C, Kuehl HS. 2016. Chimp&See: an online citizen science platform for large-scale, remote video camera trap annotation of chimpanzee behaviour, demography and individual identification. PeerJ Preprints 4:e1792v1

Abstract

The Pan African Programme: The cultured chimpanzee (PanAf) is a large-scale research project across the chimpanzee (Pan troglodytes) range which aims to better understand and model the socioecological and demographic drivers of chimpanzee diversity. As part of the PanAf, over 350,000 1-minute camera trap videos have been recorded. To annotate this large data set and ascertain individual chimpanzee identifications from 39 different temporary and collaborative chimpanzee research sites, we developed the web-based citizen science platform Chimp&See (www.chimpandsee.org) in collaboration with the Zooniverse. Chimp&See allows members of the general public to view the PanAf videos online and annotate which species are present and the behaviours they exhibit in each video. These citizen scientists also watch and discuss videos to determine unique chimpanzee individuals and match them from different video clips. Each video is viewed by up to 15 unique users, allowing us to obtain a confidence score based on the number of consensus matches for each identification. In this poster, we compare the accuracy and efficiency achieved by the general public on this platform to automated facial detection software and expert scientific annotators. We also evaluate whether citizen science and video camera trapping is a way forward for assessing chimpanzee age/sex structure, density and community size in a cost and time effective manner. Finally, we discuss the balance between maintaining user engagement and obtaining detailed and accurate scientific data from citizen scientists.

Author Comment

This is an abstract which has been accepted for the "Chimpanzees in Context" symposium”.