The Herbarium 2021 Half–Earth Challenge Dataset and Machine Learning Competition

Research output: Contribution to journalJournal articleResearchpeer-review


  • Fulltext

    Final published version, 3.81 MB, PDF document

  • Riccardo de Lutio
  • John Y. Park
  • Kimberly A. Watson
  • Stefano D'Aronco
  • Jan D. Wegner
  • Jan J. Wieringa
  • Melissa Tulig
  • Richard L. Pyle
  • Timothy J. Gallaher
  • Gillian Brown
  • Gordon Guymer
  • Andrew Franks
  • Dhahara Ranatunga
  • Yumiko Baba
  • Belongie, Serge
  • Fabián A. Michelangeli
  • Barbara A. Ambrose
  • Damon P. Little

Herbarium sheets present a unique view of the world's botanical history, evolution, and biodiversity. This makes them an all–important data source for botanical research. With the increased digitization of herbaria worldwide and advances in the domain of fine–grained visual classification which can facilitate automatic identification of herbarium specimen images, there are many opportunities for supporting and expanding research in this field. However, existing datasets are either too small, or not diverse enough, in terms of represented taxa, geographic distribution, and imaging protocols. Furthermore, aggregating datasets is difficult as taxa are recognized under a multitude of names and must be aligned to a common reference. We introduce the Herbarium 2021 Half–Earth dataset: the largest and most diverse dataset of herbarium specimen images, to date, for automatic taxon recognition. We also present the results of the Herbarium 2021 Half–Earth challenge, a competition that was part of the Eighth Workshop on Fine-Grained Visual Categorization (FGVC8) and hosted by Kaggle to encourage the development of models to automatically identify taxa from herbarium sheet images.

Original languageEnglish
Article number787127
JournalFrontiers in Plant Science
Pages (from-to)1-15
Publication statusPublished - 1 Feb 2022

Bibliographical note

Funding Information:
We would like to thank Kiat Chuan Tan from Google and the team at Kaggle (Walter Reade and Maggie Demkin) for their generous support in making this challenge possible. We would also like to thank everyone who entered the Herbarium 2021 Half?Earth Challenge. We are particularly grateful to the teams that provided detailed information on the model architectures and training strategies behind their winning submissions.

Funding Information:
This work was partially funded by National Science Foundation (USA) grant DEB 2054684.

Publisher Copyright:
Copyright © 2022 de Lutio, Park, Watson, D'Aronco, Wegner, Wieringa, Tulig, Pyle, Gallaher, Brown, Guymer, Franks, Ranatunga, Baba, Belongie, Michelangeli, Ambrose and Little.

    Research areas

  • datasets, fine-grained visual categorization, herbarium specimen image, hierarchical classification, machine learning competition

Number of downloads are based on statistics from Google Scholar and

No data available

ID: 300985173