Fundamental Methodologies for SLAM and 3D Reconstruction
Senior Lecturer with the Australian National University (ANU) and Research Fellow with the ARC Centre of Excellence for Robotic Vision
– All Welcome –
Probabilistic methods have been extensively applied in robotics and computer vision to handle noisy perception of the environment and the inherent uncertainty in the estimation. There are a variety of solutions to the estimation problems in today’s literature. Filtering and maximum likelihood estimation (MLE) are among the most used in robotics. In general, the existing efficient solutions provide only an estimation of the mean state vector, the resulting covariance being computationally too expensive to recover. Nevertheless, in many simultaneous robot localisation and mapping (SLAM) applications or 3D reconstruction of the environment from images, knowing only the mean vector is not enough. Data association, obtaining reduced state representations, active decisions and next best view are only a few of the applications that require fast state covariance recovery. Furthermore, in active computer vision and robotic problems, the state is updated with new observations and recomputed every step and its size is continuously growing, therefore, the estimation process may become highly computationally demanding.
This talk will discuss efficient solutions to the SLAM and 3D reconstruction problems which fully benefit from the incremental nature of the applications, and provide efficient estimation of both the mean and the covariance of the estimate. The talk will conclude with recent trends in this direction and future directions.
Viorela is a senior lecturer with the Australian National University (ANU) and research fellow with the ARC Centre of Excellence for Robotic Vision. Her research interests span from robot vision to advanced techniques for simultaneous localization and mapping (SLAM) and 3D reconstruction based on cutting-edge computational tools such as graphical models, modern optimization methods and information theory.
She received the Engineering degree in Industrial Engineering and Automation from the Technical University of Cluj-Napoca, Romania, in 2000 and the Ph.D. in Information Technologies from the University of Girona, Spain, in 2005. After the PhD studies, she joined the Robotics group at the Institut de Robótica i Informàtica Industrial, Barcelona, Spain. In 2009, she was awarded the MICINN/FULBRIGHT post-doctoral fellowship which allowed her to join the group of Prof. Frank Dellaert at College of Computing, Georgia Tech, Atlanta US. In 2010, she joined the robotic group at LAAS-CNRS, Toulouse, France to work in the ROSACE project founded by RTRA-STAE. Between 2012 and 2014 she was a research scientist at Brno University of Technology in Czech Republic.
For more information, please contact Prof Chun Wang