Researcher – Centre de coopération internationale en recherche agronomique pour le développement (CIRAD)
Hervé Goëau is a computer scientist at CIRAD, the French agricultural research and international cooperation organization working for the sustainable development of tropical and Mediterranean regions, and member of the AMAP joint research unit at Montpellier in France, an interdisciplinary laboratory conducting primary and advanced applied research on plants and plant communities. His topics of interests include deep learning and large-scale fine-grained classification applied to living species identification, biodiversity informatics, ecological monitoring, scientific data management, crowdsourcing & citizen sciences. For the last ten years, he has been one of the leading researchers involved in the development of Pl@ntNet, the citizen science project renowned for its mobile plant identification apps daily used by dozens of thousands of users in the world. He annually co-organizes, within the international CLEF Initiative (Conference and Labs of the Evaluation Forum), the LifeCLEF Plant species Identification task, an ambitious challenge with several thousand species, based on the visual content and metadata analyses, that promotes machine learning and computer vision in the biodiversity informatics field. He currently works on advanced applications of machine learning to automated weed detection in cultivated fields, plant disease identification, organ detection and measurements in herbarium collections, and on a global flora plant species identification system based on millions of pictures that require distributed deep learning on GPU clusters and HPC infrastructures.