AUTHOR=Doornweerd Jan Erik , Kootstra Gert , Veerkamp Roel F. , Ellen Esther D. , van der Eijk Jerine A. J. , van de Straat Thijs , Bouwman Aniek C. 

TITLE=Across-Species Pose Estimation in Poultry Based on Images Using Deep Learning

JOURNAL=Frontiers in Animal Science

VOLUME=Volume 2 - 2021

YEAR=2021

URL=https://www.frontiersin.org/journals/animal-science/articles/10.3389/fanim.2021.791290

DOI=10.3389/fanim.2021.791290

ISSN=2673-6225

ABSTRACT=Animal pose-estimation networks could enable animal breeders to quantitatively assess the gait and poses repeatedly on a large number of animals. However, the success of pose-estimation networks depends in part on the availability of data to learn the representation of key bodypoints. Especially with animals, data collection is not always easy, and data annotation is laborious and time-consuming. The available data is therefore often limited, but data from other species might be useful, either by itself or in combination with the target species. In this study, the across-species performance of animal pose-estimation networks and the performance of an animal pose-estimation network trained on multi-species data (turkeys and broilers) were investigated. Video recordings were made from broilers and turkeys during a walkway test representative of the situation in practice. Two single-species and one multi-species model were trained using DeepLabCut and tested on two single-species test sets. Overall, the within-species models outperformed the multi-species model, and the models applied across species, as shown by a lower raw pixel error, normalized pixel error, and higher percentage of keypoints remaining (PKR). The multi-species model had slightly higher errors with a lower PKR than the within-species models but had less than half the number of annotated frames available from each species. Compared to the single-species broiler model, the multi-species model was able to achieve lower errors for the head, left foot, and right knee keypoints, though with lower PKR. Across species, keypoint predictions resulted in high errors and low to moderate PKRs and are unlikely to be of direct use for pose and gait assessments. A multi-species model could potentially reduce annotation needs without a large impact on performance for pose assessment, however, with the recommendation to only be used if the species are comparable. If a single-species model exists it could be used as a pre-trained model for training a new model, and possibly require a limited amount of new data. Future studies should investigate the actual accuracy needed for pose and gait assessments and estimate genetic parameters for the new phenotypes before pose-estimation networks can be applied in practice.