We are so very happy to announce our latest peer-reviewed Chimp&See paper has been published in Remote Sensing in Ecology and Evolution titled “Highly precise community science annotations of video camera-trapped fauna in challenging environments”
https://zslpublications.onlinelibrary.wiley.com/doi/10.1002/rse2.402
Even though they require more time for annotation as compared to stills, camera trap videos provide valuable acoustic and visual information on species behaviour and can improve identification of individuals.
Using the Community Science (CS) platform Chimp&See (chimpandsee.org), we found that when comparing classification of 13,531 one-minute video clips done by a professional ecologist to those done by 1,744 Chimp&See community scientists, the overall, annotation precision was between 95.4% and 98.1%. Further, citizen scientists provide a wealth of additional information through additional annotations via hashtags in discussion boards.
Our annotated video data set is already being used for AI algorithm training and our findings demonstrate the competence of community scientists with challenging videos, and hold great promise for future studies on animal behaviour and population monitoring.
And a MASSIVE thank you and panthoot to our amazing mod @Eweforia (Carol) who made this incredible video abstract of our paper.
https://www.youtube.com/watch?v=SQiuAICI-yg
Finally, thank YOU for all your contributions to Chimp&See!!! Since it’s the summer, we have seen a little dip in classifications. We are about 65% through the current site, if you haven’t visited us in a while, please come to ChimpandSee.org today and help us complete the current site! Even 5 or 10 classifications a day help tremendously!
Thank you!
Mimi and the Chimp&See Team
