YOLO-Behaviour: A simple, flexible framework to automatically quantify animal behaviours from videos
Authors:
- Alex Hoi Hang ChanACAlex Hoi Hang Chan
- Prasetia PutraPPPrasetia Putra
- Harald SchuppHSHarald Schupp
- Johanna KöchlingJKJohanna Köchling
- Jana StraßheimJSJana Straßheim
- Britta RennerBRBritta Renner
- Julia SchroederJSJulia Schroeder
- William D. PearseWPWilliam D. Pearse
- Shinichi NakagawaSNShinichi Nakagawa
- Terry BurkeTBTerry Burke
- Michael GriesserMGMichael Griesser
- Andrea MeltzerAMAndrea Meltzer
Preprint
Abstract
Manually coding behaviours from videos is essential to study animal behaviour but it is labour-intensive and susceptible to inter-rater bias and reliability issues. Recent developments of computer vision tools enable the automatic quantification of behaviours, supplementing or even replacing manual annotations. However, widespread adoption of these methods is still limited, due to the lack of annotated training datasets and domain-specific knowledge required to optimize these models for animal research. Here, we present YOLO-Behaviour, a flexible framework for identifying visually distinct behaviours from video recordings. The framework is robust, easy to implement, and requires minimal manual annotations as training data. We demonstrate the flexibility of the framework with case studies for event-wise detection in house sparrow nestling provisioning, Siberian jay feeding, human eating behaviours, and frame-wise detections of various behaviours in pigeons, zebras, and giraffes. Our results show that the framework reliably detects behaviours accurately, and retrieve comparable accuracy metrics to manual annotation. However, metrics extracted for event-wise detection were less correlated with manual annotation, and potential reasons for the discrepancy between manual annotation and automatic detection are discussed. To mitigate this problem, the framework can be used as a hybrid approach of first detecting events using the pipeline and then manually confirming the detections, saving annotation time. We provide detailed documentation and guidelines on how to implement the YOLO-Behaviour framework, for researchers to readily train and deploy new models on their own study systems. We anticipate the framework can be another step towards lowering the barrier of entry for applying computer vision methods in animal behaviour.Keywords
This publication has 45 references indexed in Scilit:
- 3D-MuPPET: 3D Multi-Pigeon Pose Estimation and TrackingInternational Journal of Computer Vision, 2024
- PanAf20K: A Large Video Dataset for Wild Ape Detection and Behaviour RecognitionInternational Journal of Computer Vision, 2024
- LabGym: Quantification of user-defined animal behaviors using learning-based holistic assessmentCell Reports Methods, 2023
- Emerging technologies for behavioral research in changing environmentsTrends in Ecology & Evolution, 2022
- Deep learning tools for the measurement of animal behavior in neuroscienceCurrent Opinion in Neurobiology, 2020
- Automated video analysis of sow nursing behavior based on fully convolutional network and oriented optical flowComputers and Electronics in Agriculture, 2019
- A Guideline of Selecting and Reporting Intraclass Correlation Coefficients for Reliability ResearchJournal of Chiropractic Medicine, 2016
- Automated image-based tracking and its application in ecologyTrends in Ecology & Evolution, 2014
- Observer bias in animal behaviour research: can we believe what we score, if we score what we believe?Animal Behaviour, 2014
- Microsoft COCO: Common Objects in Context2014