Unifying Perspectives in Computational and Robot Vision

ISBN-13: 9781441945358 / Angielski / Miękka / 2010 / 206 str.

Danica Kragic; Ville Kyrki

Opis

Currently there is a gap between the research conducted in computer vision and robotics communities. There are many characteristics in common in computer vision research and vision research in robotics. Despite having these common interests, however, "pure" computer vision has seen significant theoretical and methodological advances during the last decade which many of the robotics researchers are not fully aware of. On the other hand, the manipulation and control capabilities of robots as well as the range of application areas have developed greatly. In robotics, vision can not be considered an isolated component, but it is instead a part of a system resulting in an action. Thus, in robotics the vision research must include consideration of the control of the system, in other words, the entire perception-action loop. This requires that a holistic system approach is useful and could provide significant advances in this application domain. Assembled here is a collection of some of the state of the art methods that are using computer vision and machine learning techniques and apply them in robotic applications.

Recenzje

From the reviews:

"'From Features to Actions: Unifying Perspectives in Computational and Robot Vision' was a workshop organized by the editors ... . This book is a collection of interesting research papers presented at that workshop. ... this book provides an overview of the latest research on vision algorithms, including their use and evaluation with robotic applications. ... I believe that this book will be a valuable resource for those interested in robotic vision research." (George K. Adam, ACM Computing Reviews, December, 2008)

Spis treści

Recent Trends in Computational and Robot Vision.- Extracting Planar Kinematic Models Using Interactive Perception.- People Detection Using Multiple Sensors on a Mobile Robot.- Perceiving Objects and Movements to Generate Actions on a Humanoid Robot.- Wald's Sequential Analysis for Time-constrained Vision Problems.- Pose Estimation and Feature Tracking for Robot Assisted Surgery with Medical Imaging.- A Sliding Window Filter for Incremental SLAM.- Topological and Metric Robot Localization through Computer Vision Techniques.- More Vision for SLAM.- Maps, Objects and Contexts for Robots.- Vision-Based Navigation Strategies.- Image-Based Visual Servoing with Extra Task Related Constraints in a General Framework for Sensor-Based Robot Systems.

Blurb

The field of computer vision has developed significantly over the last several years and the application areas span a broad range from medicine to robotics and surveillance.

Unifying Perspectives in Computational and Robot Vision provides an overview of the latest research surrounding vision algorithms and their use and evaluation with robotic applications.

In robotics, computer vision is mainly seen as another sensory modality and new algorithms are commonly developed for different applications. In the computer vision community, there have been more requirements for benchmarking and comparison of methods but very few are still applied in realistic settings and in real-time which is one of the strongest requirements in robotic applications. Unifying Perspectives in Computational and Robot Vision brings the communities of robot and computer
vision to a joint appreciation of the value of systems, where there is a need to consider all aspects from perception to action generation.

Based on a workshop "From Features to Actions: Unifying Perspectives in Computational and Robot Vision" that was organized by the editors at the IEEE International Conference on Robotics and Automation (Rome, April 2007), Unifying Perspectives in Computational and Robot Vision will be a valuable resource for readers.