Toward the automated detection of behavioral changes associated with the post-weaning transition in pigs

This dataset is associated with the paper "Toward the automated detection of behavioral changes associated with the post-weaning transition in pigs". This dataset comprises two folders. <br> <br> - The first folder comprises a set of annotated frame samples that were labeled by an expert animal behavioral scientist. These annotations were created to establish a representative dataset that is similar to commercial farms, and includes various scenarios of pigs in close proximity, such as when feeding or drinking, and in various lighting conditions.The augmented dataset used for training, which includes saturation, flipping, shearing, etc., is also provided. <br> To facilitate the use of this dataset in different platforms, the annotations are provided in different file formats, including those that are compatible with TensorFlow. This dataset will be valuable for researchers and practitioners who are interested in developing automated systems for detecting changes in pig behavior. <br> <br> - The second folder contains video recordings of pigs that were selected for the study, recorded between 10:00 and 14:00 h on days 2, 4, and 6 post-weaning. These particular days are associated with the digestive disruptions that accompany weaning and the transition from milk to solid feed. <br> <br>