GeoLifeCLEF2023

Overview

Continuously predicting the composition of plant species and its change in space and time at a fine resolution is useful for many scenarios related to biodiversity management and conservation, the improvement of species identification and inventory tools and for educational purposes.

The objective of this challenge is to predict the set of plant species present in a given location and time using various possible predictors: satellite images, time series (e.g. of climate variables), and other rasterized environmental data: land cover, human footprint, bioclimatic and soil variables.

To do so, we provide a large-scale training set of about 5M plant occurrences in Europe (single label data) as well as a validation set of about 5K plots with all the present species (multi-label data).

The test set includes 20K plots for which all species present must be predicted (multi-label classification).

The difficulties of the challenge are as follows: multi-label learning from single positive labels, strong class imbalance, multi-modal learning, large-scale.

Competition

• Start Date: Early March

• End Date: 17th of May

• Kaggle URL: https://www.kaggle.com/competitions/geolifeclef-2023-lifeclef-2023-x-fgvc10/overview

Organizers

• Christophe Botella, INRIA, LIRMM, Montpellier [christophe.botella@inria.fr]

• Alexis Joly, INRIA, LIRMM, Montpellier [alexis.joly@inria.fr]

• Diego Marcos Gonzalez, INRIA, Montpellier [diego.marcos@inria.fr]