The Herbarium 2022: Flora of North America is a part of a project of the New York Botanical Garden funded by the National Science Foundation to build tools to identify novel plant species around the world. The dataset strives to represent all known vascular plant taxa in North America, using images gathered from 60 different botanical institutions around the world.
The Herbarium 2022: Flora of North America dataset comprises 1.05 M images of 15,500 vascular plants, which constitute more than 90% of the taxa documented in North America. Our dataset is constrained to include only vascular land plants (lycophytes, ferns, gymnosperms, and flowering plants).
Our dataset has a long-tail distribution. The number of images per taxon is as few as seven and as many as 100 images. Although more images are available, we capped the maximum number in an attempt to ensure sufficient but manageable training data size for competition participants.
There are approximately 3,000 herbaria worldwide and they are massive repositories of plant diversity data. These collections not only represent a vast amount of plant diversity, but since herbarium collections include specimens dating back hundreds of years, they provide snapshots of plant diversity through time. Plant specimens are maintained in herbaria by pressing, drying, and mounting them; a specimen collected nearly two hundred years ago by Darwin looks much the same as one collected a month ago by an NYBG botanist. All specimens not only maintain their morphological features but also include collection dates and locations, their reproductive state, and the name of the person who collected the specimen. This information, multiplied by millions of plant collections, provides the framework for understanding plant diversity on a massive scale and learning how it has changed over time.
There are approximately 400,000 known vascular plant taxa with an estimated 80,000 still to be discovered. Herbaria contain an overwhelming amount of unnamed and new specimens, and with the threats of climate change, we need new tools to quicken the pace of species discovery. This is more pressing today as a United Nations report indicates that more than one million species are at risk of extinction, and amid this dire prediction is a recent estimate that suggests plants are disappearing more quickly than animals.
In botany, a ‘flora’ is a complete account of the plants found in a geographic region. The dichotomous keys and detailed descriptions of diagnostic morphological features contained within a flora are used by botanists to determine which names to apply to plant specimens. This year's competition dataset aims to encapsulate the flora of North America so that we can test the capability of artificial intelligence to replicate this traditional tool —a crucial first step to harnessing AI’s potential botanical applications.
This is an FGVC competition hosted as part of the FGVC9 workshop at CVPR 2022 and sponsored by NYBG.
Details of this competition are mirrored on the github page. Please post in the forum or open an issue if you have any questions or problems with the dataset.
The images are provided by the New York Botanical Garden and 59 other institutions around the world.
