Species196: A One-Million Semi-supervised Dataset for Fine-grained Species Recognition
This dataset addresses the need for better fine-grained recognition in invasive species classification, which has social and ecological value, but it is incremental as it primarily provides a new resource rather than a novel method.
The authors tackled the problem of limited scale and narrow species coverage in existing invasive species datasets by introducing Species196, a large-scale semi-supervised dataset with 196 categories, including over 19K labeled and 1.2M unlabeled images, and they benchmarked it across four experimental settings.
The development of foundation vision models has pushed the general visual recognition to a high level, but cannot well address the fine-grained recognition in specialized domain such as invasive species classification. Identifying and managing invasive species has strong social and ecological value. Currently, most invasive species datasets are limited in scale and cover a narrow range of species, which restricts the development of deep-learning based invasion biometrics systems. To fill the gap of this area, we introduced Species196, a large-scale semi-supervised dataset of 196-category invasive species. It collects over 19K images with expert-level accurate annotations Species196-L, and 1.2M unlabeled images of invasive species Species196-U. The dataset provides four experimental settings for benchmarking the existing models and algorithms, namely, supervised learning, semi-supervised learning, self-supervised pretraining and zero-shot inference ability of large multi-modal models. To facilitate future research on these four learning paradigms, we conduct an empirical study of the representative methods on the introduced dataset. The dataset is publicly available at https://species-dataset.github.io/.