LifeCLEF 2018

17 Janvier 2018

---

Catégorie : Compétitions et challenges

---

Building accurate knowledge of the identity, the geographic distribution and the evolution of living species is essential for a sustainable development of humanity as well as for biodiversity conservation. Unfortunately, such basic information is often only partially available for professional stakeholders, teachers, scientists and citizens, and often incomplete for ecosystems that possess the highest diversity. A noticeable cause and consequence of this sparse knowledge is that identifying living plants or animals is usually impossible for the general public, and often a difficult task for professionals, such as farmers, fish farmers or foresters and even also for the naturalists and specialists themselves. This taxonomic impediment was actually identified as one of the main ecological challenges to be solved during Rio’s United Nations Conference in 1992. In this context, an ultimate ambition is to set up innovative information systems relying on the automated identification and understanding of living organisms as a mean to engage massive crowds of observers and boost the production of biodiversity and agro-biodiversity data.

The LifeCLEF 2018 lab proposes three data-oriented challenges related to this vision, in the continuity of the previous editions of the lab, but with several consistent novelties intended to push the boundaries of the state-of-the-art in several research directions at the frontier of information retrieval, machine learning and knowledge engineering.

Challenge 1 - GeoLifeCLEF: Location-Based Species Recommendation

The aim of this task is to predict the list of species that are the most likely to be observed at a given location. Therefore, we will provide a large training set of species occurrences, each occurrence being associated to a tensor characterizing the local environment (local maps of temperature, precipitation, soil type, land cover, etc). The originality of GeoLifeCLEF is to transpose classical niche modeling approaches used in ecology into a (multichannel) image-based classification task.

Challenge 2 - BirdCLEF: Audio-based Bird Identification

Two challenges will be evaluated within this task: (i) the identification of a particular bird specimen in a recording of it, and (ii), the recognition of all specimens singing in a long sequence (up to one hour) of raw soundscapes that can contain tens of birds singing simultaneously. Both tasks will rely on the same training set containing about 50K recordings of 1500 species of central and south America (the largest bioacoustic dataset in the literature).

Challenge 3 - ExpertLifeCLEF: Experts vs. Machines plant identification

The objective of the challenge will be to evaluate how far automated identification systems are from the human expertise. Therefore a very large training set of more than 1 million images illustrating 10K plant species will be provided. Then, the goal of the task will be to return the most likely species for each observation of a test set provided without the species names. A small fraction of this test set identified by a pool of highly skilled experts will be used to conduct the experts vs. machines evaluation.

The participants will have a chance to submit a paper describing their system, which will be published in the CLEF Labs Working Notes. Furthermore, the groups of the best performing systems will be invited to give an oral presentation at CLEF 2018 and others will be given the option of presenting a poster.

Schedule

• Jan 2018: training data release
• March - April 2018: test data release
• 1st May 2018: deadline for submission of runs by the participants
• 15th May 2018: release of processed results by the task organizers
• 15th June 2018: deadline for submission of working notes papers by the participants
• 10th-14th Sept 2018: CLEF 2018 Avignon, France

Registration

LifeCLEF relies on the CrowdAI open platform for registration, submission and results management. Participants have to first create a crowdAI account (https://www.crowdai.org/participants/sign_up), and then go to the Challenges tab to access a specific LifeCLEF challenge. In order to be compliant with the CLEF registration requirements, please edit your profile by providing the following information : First name, Last name, Affiliation, Address, City. Regarding the username, please choose a name that represents your team/organism since it will be used for reference purposes.