RGB-D depth cameras have the potential to become a key component for solving real-world problems. With the drop of sensor prices, they have become a commercial success and their popularity in the research community increased. Although many publications appeared in the last years, they are spread over a variety of conferences and workshops on computer vision, robotics, human-computer interaction, and augmented reality. This makes it difficult to assess the impact of RGB-D depth cameras and the progress in this field. The proposed tutorial intends to discuss the basics, underlying principles and cutting-edge results of a comprehensive list of topics in RGB-D perception.

Schedule:

09:15-10:15 Introduction to RGB-D cameras (slides)

10:15-10:45 Coffee break

10:45-12:00 Pose estimation of humans and cameras (slides)

12:00-13:30 Lunch (provided)

13:30-14:15 Recognition and scene understanding (slides)

14:15-15:00 Face analysis (slides, manuscript)

15:00-15:30 Coffee break

15:30-16:00 Action recognition (slides, manuscript)

16:00-17:00 Commercial SDKs and APIs (slides part1, slides part1, slides part2)