PolarBearVidID: A Video-Based Re-Identification Benchmark Dataset for Polar Bears

Simple Summary Zoos use automated systems to study animal behavior. These systems need to be able to identify animals from different cameras. This can be challenging, as individuals of the same species might look very alike. AI is the best way to automatically perform this task, especially when using videos instead of images because they show the animal's movement as additional information. To train the AI model, one needs to have data. This study introduces a new dataset called PolarBearVidID that includes video sequences of 13 polar bears in various poses and lighting conditions. Our AI model is able to identify them with 96.6% accuracy. This shows that using the animals' movements can help identify them. Automated monitoring systems have become increasingly important for zoological institutions in the study of their animals' behavior. One crucial processing step for such a system is the re-identification of individuals when using multiple cameras. Deep learning approaches have become the standard methodology for this task. Especially video-based methods promise to achieve a good performance in re-identification, as they can leverage the movement of an animal as an additional feature. This is especially important for applications in zoos, where one has to overcome specific challenges such as changing lighting conditions, occlusions or low image resolutions. However, large amounts of labeled data are needed to train such a deep learning model. We provide an extensively annotated dataset including 13 individual polar bears shown in 1431 sequences, which is an equivalent of 138,363 images. PolarBearVidID is the first video-based re-identification dataset for a non-human species to date. Unlike typical human benchmark re-identification datasets, the polar bears were filmed in a range of unconstrained poses and lighting conditions. Additionally, a video-based re-identification approach is trained and tested on this dataset. The results show that the animals can be identified with a rank-1 accuracy of 96.6%. We thereby show that the movement of individual animals is a characteristic feature and it can be utilized for re-identification.