NESTLER - Poultry Behaviour Analytics Detection Dataset

This dataset was created under NESTLER (oNe hEalth SusTainabiLity partnership between EU-AFRICA for food sEcuRity) Horizon Research Project, Funded by the European Union under Grant Agreement no.101060762.

The dataset contains videos of poultry in their pen along with bounding box and body keypoints annotations. The bounding box annotations identify and localize chicken out of four categories which are "Walking", "Pecking", Standing", "Lying". The body keypoints detect four body parts (keypoints) out of each chicken. 

The aim of the dataset is to help extract useful knowledge regarding the poultry behaviour and assess their health status or detect any abnormality.

Three of the videos ( job_000004.mp4, job_000007.mp4, job_000008.mp4) were captured at Silven, Bulgaria at a poultry pen pilot under the supervision of RINISOFT and the other three videos ( job_000009.mp4, job_000010.mp4, job_000011.mp4) were captured at Muhanba, Rwanda at a poultry pen pilot under the supervision of RAB.