[Paper Review] LifeCLEF Plant Identification Task 2014

Hervé Goëau, Pierre Bonnet+1 more|ArXiv.org|Sep 28, 2025

Genomics and Phylogenetic Studies41 references78 citations

TL;DR

A comprehensive overview of the LifeCLEF plant identification challenge conducted in 2015, evaluating methods on over 100K images across 1000 West European plant species using participatory sensing data from a large contributor base.

ABSTRACT

The LifeCLEF plant identification challenge aims at evaluating plant identification methods and systems at a very large scale, close to the conditions of a real-world biodiversity monitoring scenario. The 2015 evaluation was actually conducted on a set of more than 100K images illustrating 1000 plant species living in West Europe. The main originality of this dataset is that it was built through a large-scale participatory sensing plateform initiated in 2011 and which now involves tens of thousands of contributors. This overview presents more precisely the resources and assessments of the challenge, summarizes the approaches and systems employed by the participating research groups, and provides an analysis of the main outcomes.

Motivation & Objective

• Assess plant identification methods at a large scale under realistic biodiversity monitoring conditions.
• Describe the dataset built via a participatory sensing platform with tens of thousands of contributors.
• Summarize the approaches and systems used by participating research groups.
• Analyze the outcomes and insights from the 2015 evaluation.

Proposed method

• Describe the LifeCLEF plant identification dataset and evaluation setup.
• Summarize participating teams' approaches and system architectures.
• Analyze performance outcomes and key findings of the competition.

Experimental results

Research questions

• RQ1 How well can plant identification methods scale to over 100K images and 1000 species?
• RQ2 What are the strengths and weaknesses of participating approaches in a real-world biodiversity monitoring scenario?
• RQ3 What insights emerge from the 2015 LifeCLEF plant identification evaluation?

Key findings

• The dataset comprises more than 100K images of 1000 plant species from West Europe.
• The dataset was built through a large-scale participatory sensing platform initiated in 2011 with tens of thousands of contributors.
• The overview presents resources, assessments, approaches of participating groups, and analysis of main outcomes.