10:00 Opening
10:10 Keynote 1
Davide Scaramuzza, University of Zurich, Switzerland
Agile, Autonomous, Vision-based Drones
abstract
bio
Autonomous quadrotors will soon play a major role in search-and-rescue, delivery, and inspection missions, where a fast response is crucial. However, their speed and maneuverability are still far from those of birds and human pilots. Agile flight is particularly important: since drone battery life is usually limited to 20-30 minutes, drones need to fly faster to cover longer distances. However, to do so, they need faster sensors and algorithms. Human pilots take years to learn the skills to navigate drones. What does it take to make drones navigate as good or even better than human pilots? Autonomous, agile navigation through unknown, GPS-denied environments poses several challenges for robotics research in terms of perception, planning, learning, and control. In this talk, I will show how the combination of both model-based and machine learning methods united with the power of new, low-latency sensors, such as event cameras, can allow drones to achieve unprecedented speed and robustness by relying solely on onboard computing.
Davide Scaramuzza is a Professor of Robotics and Perception at both departments of Informatics (University of Zurich) and Neuroinformatics (joint between the University of Zurich and ETH Zurich), where he directs the Robotics and Perception Group. His research lies at the intersection of robotics, computer vision, and machine learning, using standard cameras and event cameras, and aims to enable autonomous, agile navigation of micro drones in search and rescue applications. After a Ph.D. at ETH Zurich (with Roland Siegwart) and a postdoc at the University of Pennsylvania (with Vijay Kumar and Kostas Daniilidis), from 2009 to 2012, he led the European project sFly, which introduced the PX4 autopilot and pioneered visual-SLAM-based autonomous navigation of micro drones in GPS-denied environments. From 2015 to 2018, he was part of the DARPA FLA program (Fast Lightweight Autonomy) to research autonomous, agile navigation of micro drones in GPS-denied environments. In 2018, his team won the IROS 2018 Autonomous Drone Race, and in 2019 it ranked second in the AlphaPilot Drone Racing world championship. For his research contributions to autonomous, vision-based, drone navigation and event cameras, he won prestigious awards, such as a European Research Council (ERC) Consolidator Grant, the IEEE Robotics and Automation Society Early Career Award, an SNSF-ERC Starting Grant, a Google Research Award, the KUKA Innovation Award, two Qualcomm Innovation Fellowships, the European Young Research Award, the Misha Mahowald Neuromorphic Engineering Award, and several paper awards. He co-authored the book "Introduction to Autonomous Mobile Robots" (published by MIT Press; 10,000 copies sold) and more than 100 papers on robotics and perception published in top-ranked journals (Science Robotics, TRO, T-PAMI, IJCV, IJRR) and conferences (RSS, ICRA, CVPR, ICCV, CORL, NeurIPS). He has served as a consultant for the United Nations' International Atomic Energy Agency's Fukushima Action Plan on Nuclear Safety and several drones and computer-vision companies, to which he has also transferred research results. In 2015, he cofounded Zurich-Eye, today Facebook Zurich, which developed the visual-inertial SLAM system running in Oculus Quest VR headsets. He was also the strategic advisor of Dacuda, today Magic Leap Zurich. In 2020, he cofounded SUIND, which develops camera-based safety solutions for commercial drones. Many aspects of his research have been prominently featured in wider media, such as The New York Times, BBC News, Discovery Channel, La Repubblica, Neue Zurcher Zeitung, and also in technology-focused media, such as IEEE Spectrum, MIT Technology Review, Tech Crunch, Wired, The Verge.
video
10:40 Keynote 2
Lorenzo
Bruzzone, University of Trento, Italyabstract
bio
Multitemporal data and image time-series are a crucial information source for a large number of remote sensing applications. Current Earth Observation satellites (e.g., Sentinel-1 and Sentinel-2) can acquire huge amounts of images with very short revisit times. This advanced technological scenario is not fully supported yet by the capability of automatically extracting semantic from the data. Deep learning methodologies can play a very important role in the analysis of image time series addressing problems ranging from classification to change detection. However, compared with the large amount of existing techniques for semantic segmentation of single images, a relatively limited effort has been devoted to the definition of methodologies and architectures that properly analyze multitemporal data. This presentation will focus on this scenario pointing out both methodological and application challenges that should be addressed for a full exploitation of multitemporal satellite capabilities.
Lorenzo Bruzzone is currently a Full Professor of
telecommunications at the University of Trento where he is the
founder and the director of the Remote Sensing Laboratory (
https://rslab.disi.unitn.it/) in the Department of Information Engineering and Computer Science. His current research interests are in the areas of remote sensing, radar and SAR, signal processing, machine learning and pattern recognition. He promotes and supervises research on these topics within the frameworks of many (40+) national and international projects. Among the others, he is currently the Principal Investigator of the
Radar for icy Moon exploration (RIME) instrument in the framework of the
JUICE mission of the European Space Agency (ESA) and of the
High Resolution Land Cover project in the framework of the Climate Change Initiative of ESA. He is the author (or coauthor) of 294 publications in referred international journals, more than 340 papers in conference proceedings, and 22 book chapters. He is editor/co-editor of 18 books/conference proceedings and 1 scientific book. His papers are highly cited, as proven from the total number of citations (36000+) and the value of the h-index (88) (source: Google Scholar). He was invited as keynote speaker in many (40+) international conferences and workshops. Since 2019 he has been Vice-President for Professional Activities of the IEEE Geoscience and Remote Sensing Society (GRSS). Dr. Bruzzone is the recipient of many international awards. He is the co-founder of the IEEE International Workshop on the Analysis of Multi-Temporal Remote-Sensing Images (MultiTemp). Since 2003 he has been the Chair of the SPIE Conference on Image and Signal Processing for Remote Sensing. He has been the founder of the IEEE Geoscience and Remote Sensing Magazine for which he has been Editor-in-Chief between 2013-2017. He has been
Distinguished Speaker of the IEEE GRSS between 2012-2016. He is an IEEE Fellow.
Earth Observations and Machine Learning for Enhancing Global Agricultural Monitoring and Food Security
abstract
bio
Global food security is one of the major challenges we face in this Century. Innovation in developing robust and scalable measures to monitor the world’s crops in a timely, transparent manner is a key component in the fight for global food security. Such information on global crop distribution, crop conditions and food production prospects is indispensable for providing early warning of impending food shortages, for stabilizing food prices, anticipating trade needs, and enhancing farmer resilience, to name a few. The increasing frequency and severity of extreme weather events and more recently the ongoing COVID19 pandemic serve to further highlight critical agricultural information gaps and the urgency for addressing these at the field to global scale. Earth Observations and Artificial Intelligence technologies hold immense promise to help fill these gaps, with recent advances transforming the agricultural sector. This talk will discuss ongoing agricultural monitoring efforts using satellite data and machine learning under the NASA Harvest Program and will highlight impacts, challenges, and opportunities to realize the promise that these technologies hold for enhancing global agricultural monitoring and food security.
Dr. Becker-Reshef is the Director of NASA Harvest (NASA’s Applied Science Program on Food Security and Agriculture housed at University of Maryland), Visiting Professor at University of Strasbourg and Program Scientist at the GEOGLAM Secretariat. Her work is focused on the application of satellite information for agricultural monitoring from the field to global scales, in support of decisions in food security and agricultural markets. She worked closely with national and international partners to initiate the GEOGLAM (GEO Global Agricultural Monitoring) Program, adopted by the G20 in 2011 and within this program leads the Crop Monitor initiative which provides global, consensus-driven crop condition assessments on a monthly basis. Her background is in soil sciences and remote sensing and she earned her Ph.D in Geographical Sciences from the University of Maryland. In 2016 She was recognized by the US State Department for her work on Food Security and Technologies, winning the US Asia-Pacific Economic Cooperation (APEC) Science Prize for Innovation, Research, and Education (ASPIRE) awarded by John Holdren, Former Assistant to the President for Science and Technology.
video
15:00 Oral Session 2 - Pose, Point Clouds, Photogrammetry
Gordon Christie (JHU); Kevin Foster (JHU); Shea Hagstrom (Johns Hopkins University Applied Physics Laboratory); Gregory D. Hager (The Johns Hopkins University); Myron Brown (JHU)
Single View Geocentric Pose in the Wild
paper bibtex
@InProceedings{Christie_2021_CVPR_Workshops,
author = {Christie, Gordon and Foster, Kevin and Hagstrom, Shea and Hager, Gregory D. and Brown, Myron},
title = {Single View Geocentric Pose in the Wild},
booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops},
month = {June},
year = {2021}
}
Dawa Derksen (ESA); Dario Izzo (ESA)
Shadow Neural Radiance Fields for Multi-view Satellite Photogrammetry
paper bibtex
@InProceedings{Derksen_2021_CVPR_Workshops,
author = {Derksen, Dawa and Izzo, Dario},
title = {Shadow Neural Radiance Fields for Multi-view Satellite Photogrammetry},
booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops},
month = {June},
year = {2021}
}
Yi Wang (German Aerospace Center); Stefano Zorzi (Graz University of Technology); Ksenia Bittner (German Aerospace Center)
Machine-learned 3D Building Vectorization from Satellite Imagery
paper bibtex
@InProceedings{Wang_2021_CVPR_Workshops,
author = {Wang, Yi and Zorzi, Stefano and Bittner, Ksenia},
title = {Machine-learned 3D Building Vectorization from Satellite Imagery},
booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops},
month = {June},
year = {2021}
}
Takayuki Shinohara (Tokyo Institute of Technology); Xiu Haoyi (Tokyo Tech); Masashi Matsuoka (Tokyo Institute of Technology)
Point2color: 3D Point Cloud Colorization Using a Conditional Generative Network and Differentiable Rendering for Airborne LiDAR
paper bibtex
@InProceedings{Shinohara_2021_CVPR_Workshops,
author = {Shinohara, Takayuki and Haoyi, Xiu and Matsuoka, Masashi},
title = {Point2color: 3D Point Cloud Colorization Using a Conditional Generative Network and Differentiable Rendering for Airborne LiDAR},
booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops},
month = {June},
year = {2021}
}
15:40 Oral Session 3 - Benchmarks and Datasets
Nannan Qin (Purple Mountain Observatory, Chinese Academy of Sciences); Weikai Tan (University of Waterloo); Lingfei Ma (University of Waterloo); Dedong Zhang (University of Waterloo); Li Jonathan (Xiamen University)
OpenGF: An Ultra-Large-Scale Ground Filtering Dataset Built Upon Open ALS Point Clouds Around the World
paper bibtex
@InProceedings{Qin_2021_CVPR_Workshops,
author = {Qin, Nannan and Tan, Weikai and Ma, Lingfeiand Zhang, Dedong Jonathan, Li},
title = {OpenGF: An Ultra-Large-Scale Ground Filtering Dataset Built Upon Open ALS Point Clouds Around the World},
booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops},
month = {June},
year = {2021}
}
Adrian Boguszewski (Linux Polska); Dominik Batorski (Interdisciplinary Centre for Mathematical and Computational Modelling, University of Warsaw); Natalia Ziemba-Jankowska (Linux Polska); Tomasz Dziedzic (Linux Polska); Anna Zambrzycka (Agency for Restructuring and Modernisation of Agriculture)
LandCover.ai: Dataset for Automatic Mapping of Buildings, Woodlands, Water and Roads from Aerial Imagery
paper bibtex
@InProceedings{Boguszewskin_2021_CVPR_Workshops,
author = {Boguszewski, Adrian and Batorski, Dominik and Ziemba-Jankowska, Natalia and Dziedzic, Tomasz and Zambrzycka, Anna},
title = {LandCover.ai: Dataset for Automatic Mapping of Buildings, Woodlands, Water and Roads from Aerial Imagery},
booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops},
month = {June},
year = {2021}
}
Sagar Verma (CentraleSupelec); Akash Panigrahi (Granular AI); Siddharth Gupta (Granular AI)
QFabric: Multi-Task Change Detection Dataset
paper bibtex
@InProceedings{Verma_2021_CVPR_Workshops,
author = {Verma, Sagar and Panigrahi, Akash and Gupt, Siddharth},
title = {QFabric: Multi-Task Change Detection Dataset},
booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops},
month = {June},
year = {2021}
}
Christian Requena-Mesa (Computer Vision Group, Friedrich Schiller University Jena; DLR Institute of Data Science, Jena; Max Planck Institute for Biogeochemistry, Jena); Vitus Benson (Max-Planck-Institute for Biogeochemistry); Markus Reichstein (Max Planck Institute for Biogeochemistry, Jena; Michael Stifel Center Jena for Data-Driven and Simulation Science, Jena); Jakob Runge (Institute of Data Science, German Aerospace Center (DLR)); Joachim Denzler (Computer Vision Group, Friedrich Schiller University Jena, Germany)
EarthNet2021: A large-scale dataset and challenge for Earth surface forecasting as a guided video prediction task
paper
supp
bibtex
@InProceedings{Requena-Mesa_2021_CVPR_Workshops,
author = {Requena-Mesa, Christian and Benson, Vitus and Reichstein, Markus and Runge, Jakob and Denzler, Joachim},
title = {EarthNet2021: A large-scale dataset and challenge for Earth surface forecasting as a guided video prediction task},
booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops},
month = {June},
year = {2021}
}
16:20 Break
16:30 Keynote 4
Claire Monteleoni, University of Colorado, Boulder, USA
Deep Unsupervised Learning for Climate Informatics
abstract
bio
Despite the scientific consensus on climate change, drastic uncertainties remain. Crucial questions about regional climate trends, changes in extreme events, such as heat waves and mega-storms, and understanding how climate varied in the distant past, must be answered in order to improve predictions, assess impacts and vulnerability, and inform mitigation and sustainable adaptation strategies. Machine learning can help answer such questions and shed light on climate change. This talk will focus on our recent climate informatics research, in particular semi- and unsupervised deep learning approaches to studying rare and extreme events, and downscaling temperature and precipitation.
Claire Monteleoni is an Associate Professor, and the Associate Chair for Inclusive Excellence, in the Department of Computer Science at the University of Colorado Boulder, and the founding Editor in Chief of Environmental Data Science, a Cambridge University Press journal, launched in December 2020. She joined CU Boulder in 2018, following positions at University of Paris-Saclay, CNRS, George Washington University, and Columbia University. She completed her PhD and Masters in Computer Science at MIT and was a postdoc at UC San Diego. She holds a Bachelor’s in Earth and Planetary Sciences from Harvard. Her research on machine learning for the study of climate change helped launch the interdisciplinary field of Climate Informatics. In 2011, she co-founded the International Conference on Climate Informatics, which turned 10 years old in 2020, and has attracted climate scientists and data scientists from over 20 countries and 30 U.S. states. She gave an invited tutorial: Climate Change: Challenges for Machine Learning, at NeurIPS 2014.
video
17:00 Oral Session 4 - Representations, Image Processing, and Parameter Estimation
video
Ryan Mukherjee (JHU); Derek Rollend (JHU); Gordon Christie (JHU); Armin Hadzic (JHU/APL); Sally Matson (JHU); Anshu Saksena (JHU); Marisa Hughes (JHU)
Towards Indirect Top-Down Road Transport Emissions Estimation
paper bibtex
@InProceedings{Mukherjee_2021_CVPR_Workshops,
author = {Mukherjee, Ryan and Rollend, Derek and Christie, Gordon and Hadzic, Armin and Matson, Sally and Saksena, Anshu and Hughes, Marisa },
title = {Towards Indirect Top-Down Road Transport Emissions Estimation},
booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops},
month = {June},
year = {2021}
}
Arpit Jain (GE Global Research); Tapan Shah (GE ); Mohammed Yousefhussien (General Electric); Achalesh Pandey (General Electric)
Combining Remotely Sensed Imagery with Survival Models for Outage Risk Estimation of the Power Grid
paper bibtex
@InProceedings{Jain_2021_CVPR_Workshops,
author = {Jain, Arpit and Shah, Tapan and Yousefhussien, Mohammed and Pandey, Achalesh},
title = {Combining Remotely Sensed Imagery with Survival Models for Outage Risk Estimation of the Power Grid},
booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops},
month = {June},
year = {2021}
}
Ngoc Long Nguyen (ENS Paris-Saclay); Jérémy Anger (ENS Paris-Saclay); Axel Davy (ENS Paris-Saclay); Pablo Arias (ENS Paris-Saclay); Gabriele Facciolo (ENS Paris - Saclay)
Self-supervised multi-image super-resolution for push-frame satellite images
paper
supp
bibtex
@InProceedings{Nguyen_2021_CVPR_Workshops,
author = {Nguyen, Ngoc Long and Anger, Jérémy and Davy, Axel and Arias, Pablo and Facciolo, Gabriele},
title = {Self-supervised multi-image super-resolution for push-frame satellite images},
booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops},
month = {June},
year = {2021}
}
Vladan Stojnic (Faculty of Electrical Engineering, University of Banja Luka); Vladimir Risojevic (Faculty of Electrical Engineering, University of Banja Luka)
Self-Supervised Learning of Remote Sensing Scene Representations Using Contrastive Multiview Coding
paper bibtex
@InProceedings{Stojnic_2021_CVPR_Workshops,
author = {Stojnic, Vladan and Risojevic, Vladimir},
title = {Self-Supervised Learning of Remote Sensing Scene Representations Using Contrastive Multiview Coding},
booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops},
month = {June},
year = {2021}
}
17:40 Closing
© contains
modified Copernicus Sentinel data (2020), processed
by ESA, CC BY-SA 3.0 IGO.
Inbal
Becker-Reshef, NASA Harvest,
USA
