Computer Vision News - March 2018

We have asked one of the coordinators of the challenge to tell us about it. Henning Müller from HES-SO in Sierre (Switzerland) was kind enough to help. “ The objective of LifeCLEF - Henning explains - has been from the start to allow automatic species identification tools (from birds to plants, fish etc.) to be used for practical applications in biodiversity monitoring and for citizen science. Each year has had an increasing complexity with more classes, more complex data that requires multimodal fusion. ” LifeCLEF 2018 is actually composed by three data-oriented challenges: GeoLifeCLEF , the aim of which is to predict the list of species that are the most likely to be observed at a given location (see image at the bottom of the page). BirdCLEF , the aim of which is to help experts assess birds population on the field using new interactive identification tools, more time- efficient than direct observation; and also to assess biodiversity without or with very light user’s involvement. ExperLifeCLEF , aims at quantifying intrinsic data uncertainty resulting from expert identification and compare it to the output given by recent and very performant deep learning automated models (up to 90% classification accuracy over 10K species). 24 Challenge: LifeCLEF Challenge Every month, Computer Vision News reviews a challenge related to our field. If you do not take part in challenges, but are interested to know the new methods proposed by the scientific community to solve them, this section is for you. This month we have chosen to tell you about the LifeCLEF 2018 challenge, organized around CLEF 2018, which will be held in September in Avignon, France. The website of the challenge, with all its related resources, is here . Challenge on Automatic Species Identification Tools Computer Vision News