{"id":23,"date":"2021-01-12T13:13:49","date_gmt":"2021-01-12T13:13:49","guid":{"rendered":"https:\/\/sodalite.deib.polimi.it\/?page_id=23"},"modified":"2022-07-20T12:50:38","modified_gmt":"2022-07-20T12:50:38","slug":"welcome","status":"publish","type":"page","link":"https:\/\/environmentintelligence.deib.polimi.it\/","title":{"rendered":"Welcome to Environment Intelligence Research Group"},"content":{"rendered":"

Enviroment Intelligence is the discipline that applies Artificial Intelligence methods and tools to environmental problems.\u00a0 The Environment Intelligence Group is a team of researchers of the Dipartimento di Elettronica, Informazione e Bioingegneria (DEIB<\/a>) of Politecnico di Milano<\/a>.<\/p>\n

The aim of the research is to investigate novel applications of AI-powered Computer Vision to the resolution of environmental issues, in such domains as water management and waste management.<\/p>\n

The research work spans multiple disciplines, including Artificial Intelligence, Computer Vision, Big Data Management, Natural Resource Management, Software Engineering, Mobile Application Engineering, Gamification Design, an more.<\/p>\n

We are the creators of PeakLens<\/a>, an AI+CV powered application for engaging people in mountain exploration. Join the\u00a0community of more than 1 million PeakLens users!<\/p>\n\n\n\n
\"\"<\/a><\/td>\n\"\"<\/a><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

We released AerialWaste,<\/a> a dataset for the discovery of illegal landfill. The dataset contains +10.000 satellite images from three different sources at high resolution. The sites present 15 different waste types and 7 different waste storage modes.<\/p>\n

We are the creators of ODIN Diagnosis Tool<\/a>, an open-source framework assisting the life-cycle of ML applications (computer vision, time series, and generic classification) in the phases of data preparation, prediction performance evaluation, and error diagnosis. ODIN is agnostic to the training platform and input formats and can be extended with application- and domain-specific meta-annotations and metrics with almost no coding.<\/p>\n

Research is supported by multiple projects, funded by national and international agencies.<\/p>\n

Current and ongoing projects:<\/p>\n