Introduction to the special issue on visual understanding and applications with RGB-D cameras
The prevalence of affordable RGB-D cameras, such as Microsoft's Kinect and ASUS's Xtion Pro Live, is driving a revolution of the landscape of computer vision and its related researches. The pixel-level depth and color information provided by an RGB-D camera not only enables robust vision applications but also opens up new research problems and opportunities across a wide range of areas including 3D perception, robotics, human machine interactions, intelligent surveillance, and health care. With the depth information, many intractable computer vision problems are becoming possible, but they are by no means solved. New visual representations and paradigms are needed to effectively leverage the new sensors and develop robust vision systems. Since the first release of Microsoft Kinect sensors in November 2010, we have witnessed a rapid growth of research in visual understanding and applications with RBG-D cameras. This special issue brings together the most recent advances in this direction. There are 19 high quality papers covering various aspects of RGB-D camera based visual understanding including human action recognition, pose and gait recognition, object recognition, segmentation, tracking, sensor fusion, and novel applications. From these papers, the readers will obtain an excellent understanding on the state of the art of this new research area.