The field of plant identification and species composition prediction is rapidly advancing, driven by the increasing use of deep learning models and remote sensing data. Recent studies have shown that the performance of state-of-the-art deep learning models is now close to that of human experts, highlighting the potential for automated systems to support biodiversity monitoring and conservation efforts. A key challenge in this area is quantifying the uncertainty associated with automated identification systems and comparing it to the performance of human experts. Researchers are also exploring the use of large-scale participatory sensing platforms and citizen science initiatives to collect and annotate data for training and evaluating plant identification models. Furthermore, the integration of remote sensing data with species observations is enabling the prediction of species composition at high spatial resolution and continental scale. Notable papers in this area include:

• Overview of ExpertLifeCLEF 2018, which presents a comparison between human experts and automated systems for plant identification.
• Overview of GeoLifeCLEF 2023, which highlights the use of deep learning models and remote sensing data for species composition prediction.