|
Describe:
|
The 2022 GeoAI Martian Challenge is the very first in the series of GeoAI challenges hosted by the Cyberinfrastructure and Computational Intelligence Lab (CICI) and the Spatial Analysis Research Center (SPARC) of Arizona State University (ASU). This is a collaboration among researchers from multiple disciplines, organizations and sectors, including ASU, U.S. Geological Survey (USGS), Jet Propulsion Laboratory (JPL), Oak Ridge National Lab, Esri, Google and American Geographical Society. The goal of this object detection challenge is to broaden the participation of, but not limited to, students and early career scholars in geospatial sciences, computer science, Earth and space sciences, data science, and fields relevant to this exciting research community. Through this challenge, we also hope to form a strong research network to facilitate collaboration among researchers and practitioners from different disciplines to jointly promote cutting-edge GeoAI research and education, as well as their translational solutions to problems in a wide range of environmental and social science domains. Virtual hands-on training sessions on designing and developing AI and deep learning models will be offered during the open period of this challenge (March 2022 to September 2022), so this is more than a challenge but also a platform for developing and diversifying the next-generation (Geo)AI workforce.
This year, our focus is detecting craters on Mars’ surface, so we have named it the GeoAI Martian Challenge. This is to celebrate the anniversary of the successful landing of the Mars Perseverance rover on the red planet on February 18, 2021. We hope this event will promote the applications of cutting-edge AI technology in planetary and geospatial sciences, contribute to the exploration of future human life in space, and allow us to bring this experience back to Earth to study the changing environment of our living planet. In this object detection challenge, we have prepared Mars crater datasets that contain over 100,000 images and 300,000 craters located in nearly every corner of the Mars surface. Our goal is to encourage participants to develop novel GeoAI models for detecting the locations of craters on each image. The craters can be as small as ~100 meters in diameter or up to 2km in diameter. So, this is an interesting and challenging AI object detection problem. More information about the dataset and the details for the competition can be found here: http://cici.lab.asu.edu/martian.
This challenge is open literally to everyone who is interested in AI and geospatial sciences. Participants can be individuals or a team of AI fans. We especially encourage students, early-career scholars, women and underrepresented minorities to participate this event. We will select the top three solutions of the competition for a cash award (First prize: $500, Second prize: $300, Third prize: $200), a certificate as well as a generous gift from Esri. Selected participants will also have the opportunity to conduct internship in the CICI lab of ASU. We look forward to your participation!
|
