Pl@ntNet-300K image dataset

<p>This paper presents a novel image dataset with high intrinsic ambiguity and a long-tailed distribution built from the database of Pl@ntNet citizen observatory. It consists of 306146&nbsp;plant images covering 1081&nbsp;species. We highlight two particular features of the dataset, inherent to the way the images are acquired and to the intrinsic diversity of plants morphology:</p> <p>&nbsp; &nbsp; (i) the dataset has a strong class imbalance, i.e. a few species account for most of the images, and,</p> <p>&nbsp; &nbsp; (ii) many species are visually similar, rendering identification difficult even for the expert eye.</p> <p>&nbsp; &nbsp; These two characteristics make the present dataset well suited for the evaluation of set-valued classification methods and algorithms. Therefore, we recommend two set-valued evaluation metrics associated with the dataset (macro-average top-k accuracy&nbsp;and macro-average average-k accuracy) and we provide baseline results established by training deep neural networks using the cross-entropy loss.</p> <p>A full description of the dataset as well as baseline experiments can be found in the following&nbsp;publication:</p> <p>&quot;<a href="https://openreview.net/forum?id=eLYinD0TtIt">Pl@ntNet-300K: a plant image dataset with high label ambiguity and a long-tailed distribution</a>&quot;, Camille Garcin, Alexis Joly, Pierre Bonnet, Antoine Affouard, Jean-Christophe Lombardo, Mathias Chouet, Maximilien Servajean,&nbsp;Titouan Lorieul and Joseph Salmon, in Proc. of Thirty-fifth Conference on Neural Information Processing Systems, Datasets and Benchmarks Track, 2021.</p> <p>&nbsp;Please cite the above&nbsp;reference for any publication using the dataset.</p> <p>Utilities to load the data and train models with pytorch can be found here: <a href="https://github.com/plantnet/PlantNet-300K/">https://github.com/plantnet/PlantNet-300K/</a></p>